Download Senior Thesis Guide - Yale Sociology

Document related concepts
no text concepts found
Transcript
 Writing Sociology A Guide for Senior Theses Rebekah P. Massengill, Ph.D. Adapted for use at Yale by Celene Reynolds, M.A. and Frederick Wherry, Ph.D. Originally published as Writing Sociology: A Guide for Junior Papers and Senior Theses, Department of Sociology, Princeton University. The author has granted Yale permission for its use in 2014-­‐2015. W r i t i n g S o c i o l o g y P a g e | 2 Table of Contents Introduction ................................................................................................................................. 3 Chapter 1: The Research Question ............................................................................................. 5 Chapter 2: Defining the Importance of Your Research ............................................................. 11 Chapter 3: The Literature Review ............................................................................................. 14 Chapter 4: Finding Data ............................................................................................................ 21 Chapter 5: Methods .................................................................................................................. 26 Chapter 6: Data Analysis ........................................................................................................... 35 Chapter 7: Discussing Your Findings and Drawing Conclusions ................................................ 43 Chapter 8: Bibliography ............................................................................................................ 44 Chapter 9: Other Helpful Information ...................................................................................... 48 Appendix: Other Resources ...................................................................................................... 58 W r i t i n g S o c i o l o g y P a g e | 3 Introduction Congratulations! As you begin to read this handbook, you are initiating the process of conducting original research worthy of fulfilling the university’s independent research requirements for sociology, a senior thesis for seniors. Together with the mentorship of your faculty advisor, this handbook will help you think through the various steps of your independent research, and in doing so make the most of the relationship with the advisor who will supervise your research on behalf of the sociology department. Yale takes its independent research requirements very seriously, and undergraduates at Yale perform authentic, scholarly research that, at its best, contributes to the larger scholarly conversations taking place within various academic disciplines in the academy. In sociology in particular, independent research means a couple of distinct things. First and foremost, sociological research must be informed by a scholarly literature. Sociologists seek to better understand society and build theories that help us to make sense of and understand our social worlds. Independent research must treat seriously the established sociological literature on a given topic and make clear how the student’s research contributes to the larger literature in the field. Second, sociological research at the undergraduate level will normally involve the analysis of empirical data.1 While some students analyze existing statistical datasets, other students will conduct their own surveys, while others will use interviews or field observations to gather their data. Other sources of data might include texts, images, or archival data sources such as newspapers, legal proceedings, or organizational records. Whether using quantitative or qualitative methods, the analysis of this empirical data represents the most important element of the thesis. Rather than summarizing existing studies or reporting on the findings of others, students doing independent research in sociology seek to contribute to this larger literature by analyzing data themselves. The collection or location of appropriate data, and the original analysis and discussion of results are what make this work distinct from much of the writing you may have done in other classes. At its best, independent research in sociology asks students to contribute to a larger field of sociological knowledge and will make students familiar with the methods of careful, systematic data collection and analysis practiced within the discipline. Strong independent work in sociology will have substantial elements of originality in its conception of its subject, in the evidence and reasoning it brings to bear on that subject, and/or in the analytical techniques it employs. The best theses in sociology conduct research that would interest and inform even specialists in the student’s field of study. While students often fear that “independent” research will mean “solitary” research, this impression of sociological research couldn’t be farther from the truth! This guide frequently points you to other important and useful live resources available throughout the university. Also keep in mind that the most important resources for you in the research process are your professors and the teaching fellow (TF): talking with these people early and often is one of the best ways to ensure smooth sailing in your 1
Rarely, a student will write a senior thesis that focuses exclusively on sociological theory. If you wish to pursue this option, you should discuss it with your advisor as soon as possible to determine if this is an acceptable course of action. W r i t i n g S o c i o l o g y P a g e | 4 independent research. In this way, independent research need not be isolating – in fact, academic scholarship is an activity that can only be practiced in community. Everyone in the university – ranging from its most senior tenured professors to graduate student instructors – looks to their colleagues and collaborators for help and guidance in their research activities. Writing your thesis is simply the next step in the process of becoming a member of this academic community – a process inaugurated with your freshman year seminar, and fulfilled with the completion your senior thesis. How to Use this Guide Your thesis will contain some version of each of the following elements: 1.
2.
3.
4.
5.
6.
7.
8.
A research question A statement of why your research question is important A literature review of relevant social scientific literature Description of the data you have selected for analysis A discussion of the methods used for the data analysis Analysis of results Discussion of the significance of the analysis, including general conclusions A Bibliography of works cited Accordingly, this guide has eight chapters arranged around each of these items, as well as an additional chapter with helpful advice about developing a productive relationship with your advisor and planning your work. In the process of taking you through each of these elements required in your independent work, this handbook will also cover other important logistical issues as well, such as suggestions for different places you can find data, negotiating IRB approval issues, and protecting yourself against plagiarism. This guide also contains an appendix that lists other helpful information as well as multiple resources within the University community that you can also use for assistance with various stages of this process. If you attend carefully to each of these items, you will find that the process of independent research will be challenging, but not at all overwhelming. This guide seeks to help you navigate that process from start to finish, doing your best work all along the way. W r i t i n g S o c i o l o g y P a g e | 5 Chapter 1: The Research Question Of all the pieces of your research project, this is probably the most important – and perhaps the most challenging. You probably chose to major in sociology because you are interested in the social world and the kinds of questions that interest sociologists: questions about social inequality, the nature and functions of social groups and organizations, large-­‐scale changes in collective and national identities, and the subtle habits and cues that shape people’s interactions with one another – and this only scratches the surface! Questions in sociology often take some of the following forms: • Questions about the meaning of certain activities, practices, or experiences for particular social groups. (E.g. How do parents respond when learning that their child has been diagnosed with Sickle-­‐Cell Anemia? What does it mean to be included in a larger group of people afflicted with a disease? How does identification with a particular illness change the experience of illness and treatment for families managing the disease?) • Questions about the ways that identification with larger social categories – race, ethnicity, religion, political identification, gender – affect aspects of social life. (What’s the relationship between being religiously conservative and politically conservative? Are evangelicals more likely to vote Republican? How do ideas about masculinity influence shop floor culture for workers in the auto industry?) • Questions about the influence of particular variables on other variables or outcomes, including questions that compare groups and track trends across a broader scale. (What difference does income level have on a mother’s likelihood of receiving prenatal care during pregnancy? Are mothers who receive prenatal care less likely to drink during pregnancy than those who don’t? How might these relationships vary by age? Do clinics located in poor and nonpoor neighborhoods offer different forms of prenatal care to women?) The challenge for many student sociologists, then, is settling on a topic and finding a research question that is of an appropriate scope for analysis in an independent research project. Your question should be complex enough to warrant serious treatment in a lengthy paper, but focused enough that you can do a thorough job with your analysis. (For a discussion of suggested length and format of a thesis, see Chapter 9, “Other Helpful Information.”) In an ideal world, the process of finding your research question would go something like this: 1. Read lots of books and realize that famous sociologists have neglected a simple, but crucial research question that demands an answer. 2. Find data that offers an answer to this research question. 3. Analyze your newly discovered data, and thus provide an answer to this captivating and important research question. W r i t i n g S o c i o l o g y P a g e | 6 4. Win awards and accolades for your dazzling brilliance. In reality, of course, social research rarely works in exactly this way (although it would be nice – and a lot easier – if it did!) Instead, the process of arriving at a research question, for students and seasoned researchers alike, often looks more like this: 1. Start thinking about a topic that interests you. 2. Brainstorm lists of questions that might be worth investigating in social research. 3. Read books and articles about this topic to see what other researchers have found; return to #2 and thus revise your questions. 4. Find a data source (or sources) on your topic of interest, and think about the concrete questions raised by this data and how they correspond (or not) to what you’ve found in #3. 5. Go back to #2 . . . And you get the idea. In fact, to even number these steps seems futile, since this process is one in which the student researcher is doing a number of things at one time – brainstorming, reading, and testing the waters in thinking about how real-­‐world data might inform the kinds of questions you are asking. To illustrate what this muddier version of research might look like in practice, think about a potential paper on the broad topic of the experiences of well-­‐educated women in the professional workplace. You might be interested in a number of questions about this topic, for example: •
•
•
•
•
How do women make decisions about investing in work vs. investing in children? What difference do workplace rules and policies make in whether or not women choose to stay home after having children or continue to invest in their careers? How does women’s compensation in professional work situations compare to that of men? Do women born since 1965 act differently in their career decisions from women born between 1945 and 1965? When well-­‐educated women drop out of the labor market after having children, what reasons do they give for this decision? In working towards a more focused research question, you would want to do some reading around this issue, focusing on sociological articles and books (in other words “the literature”) about women in the labor market, work/family issues, and workplace policies and procedures. As we’ll discuss in Chapter 3, you’ll want to take lots of notes in your literature review about what other scholars consider unanswered questions or places where published studies and findings do not agree. This will help to refine your research question significantly. You would also want to think about some potential data sources that would answer these questions. For instance, if your main question of interest is about women’s pay in professional workplace situations, you would probably want to look at a quantitative dataset that you would analyze using statistical methods, such as the U.S. Census or the Current Population Study (CPS), two publicly available datasets available from locations we’ll discuss in Chapter 4. If what you’re really interested in are the reasons why people do things, or what people have to say about their own experiences (such as why W r i t i n g S o c i o l o g y P a g e | 7 well educated, professional women choose to leave the labor market in order to stay home with young children), then you could gather your own data for your senior thesis by interviewing people in the form of a qualitative sample. We will also discuss this further in Chapter 4. Although this strategy may sound daunting, it’s not as hard as it sounds. In this particular example, you might find that you could interview a sample of women who are involved in local Yale mother’s groups. In either scenario, your research questions will be developed through a process of brainstorming, consulting the literature, and generating ideas in consultation with an identifiable data source. What you are looking for, in essence, are holes in existing research: what do we still need to know about the topic that interests you, and how could you help to answer them in your independent research? Troubleshooting We all know that it happens: along the way to finding your research question, you make a wrong turn, get stuck, or have a hard time getting started. In the remainder of this chapter we will discuss some common problems in this process and some strategies you can use to overcome them. 1. Can’t find a topic. The Problem: This is everyone’s worst nightmare. You can’t decide what you’re really interested in or keep coming up with reasons why the ideas that you have won’t be feasible. Or maybe you keep on changing topics – you spend the first part of the semester planning to research segregated housing but then decide, after fall break, that you want to switch to something about politics and the media. Strategies: First of all, don’t beat yourself up: we’ve all done this to some extent. But you must intervene early and allow yourself to do the hard work of brainstorming your way out of the muddy waters of confusion. In many cases, being reluctant to choose a topic or to settle on one idea out of many is a reflection of our underlying perfectionism – when we can’t immediately find a topic that seems “perfect” and thus exciting, worthwhile, and manageable, we give up and move on to another idea. We perfectionists get more and more panicked as time ticks by and we realize that we haven’t really gotten started. The best medicine here is to think about what you are really, truly interested in. Most of what interests us as sociologists often comes from our own experience – even Weber admitted as much in his essays found in his classic book The Methodology of the Social Sciences. With that in mind, sit down with pen and paper in hand and brainstorm as many questions as you can think of about a particular topic that really and truly interest you. If you like, do this for a couple of topics and then compare the lists. If something on a list makes your eyes glisten, go for it! Even better, make an appointment with your advisor or the Teaching Fellow and take your lists to the meeting to discuss your potential ideas. W r i t i n g S o c i o l o g y P a g e | 8 2. You have a hard time moving from topic to research question. The Problem: You’ve got some great ideas about broad topics – the significance of the internet in social inequality, or the effect of a family member’s incarceration on children – but you’re having trouble arriving at a concrete research question. Strategies: Booth, Colomb, and Williams talk about this very issue in their great book The Craft of Research (University of Chicago Press, 3rd Ed, 2008). They offer a number of strategies for resolving this issue, which mostly involve breaking your topic down into smaller pieces that allow you to arrive at a focused research question. They focus on the importance of asking smaller analytical questions that focus on the how and why of research rather than the more descriptive questions of who, what, when, or where. Some of the strategies that Booth et. Al suggest include: 1. Breaking your topic down into parts and asking questions about how those parts are related. Specifically, ask yourself what smaller parts make up your topic, and then think about either how those parts of your topic relate, or how your topic is a part of a larger system. For example: TOPIC PARTS The internet and social inequality Access to computers in rural/urban areas Technological literacy and training Computer availability in schools, libraries, etc. by neighborhood Computer use by sociodemographic groups (age, race, ethnicity, language, etc.) Federal funding for technological initiatives Incarceration and children Different effects of incarceration of mother, father, sibling, etc. Effects of incarceration on family income Effects of incarceration on emotional well-­‐being Rates of incarceration among blacks vs. whites Length of incarceration and influence on children Frequency of/access to visitation for family members W r i t i n g S o c i o l o g y P a g e | 9 When you break your topic down into smaller parts, more focused research questions become apparent. For example, thinking about the different parts of incarceration and its effect on families suggests a number of more detailed, manageable research questions. For example, instead of asking “How does incarceration affect children” you might ask “How does frequency of prison visitation affect family stability during incarceration?” 2. Put your topic into historical context For our particular examples here, we could think about how internet access or computer use has changed over time for different social groups, or about how incarceration is a factor in neighborhood stability over time. What research questions does that help us generate (for example, “How has the mass incarceration of black males affected the significance of motherhood in poor, urban neighborhoods?”) 3. Compare and Contrast Think about how your topic is a snapshot that compares to a larger picture – how does computer and internet use contribute to the urban/rural divide in the present day United States? Or think about how instances of your topic are like or unlike something else – how is incarceration similar to or different from losing a parent to death or divorce? 3. You find that someone else has asked – and answered – exactly your questions. The Problem: You find, well into your research and literature review, that another scholar or scholars have explored exactly the same question that you are pursuing and thus fear you will have to start over in your research. Strategies: How you handle this depends on how soon you discover the overlap between other scholarly work and your own, developing research questions. If you find this early on in your literature review, you can easily adjust your question slightly to think about how the work you’ve discovered can actually strengthen your research. For example, does the work you’ve found suggest some unanswered questions that your research can address? Does it propose a theoretical framework than you can build off of, or further refine or substantiate? Or perhaps the other research bases its conclusions off of slightly different data than the data you are considering using – in this case your contribution can be to help build even further support for a particular claim (or perhaps show us how that claim should be refined or adjusted slightly based on new findings from your analysis). In any case, this isn’t necessarily terrible news – remember that scholarship is the work of an academic community, and to discover that your questions are already being investigated should affirm for you that you are asking analytical questions that others in the academic community deem important and valuable. Remember that “original” research need not be groundbreaking. Rather, the process of scientific research is one of building collective knowledge slowly, brick by brick. You should also, of course, use your professors and TFs to help you think about how your research can integrate existing sources into an original research question. W r i t i n g S o c i o l o g y P a g e | 10 Chapter 2: Defining the Importance of Your Research As you continue to work on developing your research question you will have to be thinking about an important, related question: why your proposed research question is important. In other words, so what? You may be terribly interested how in evangelical and Jewish students interrelate on the Yale campus, but unless you can give us a concise statement of why this research question matters for the larger enterprise of sociology your research will be incomplete. In the Writing Seminar you took as a freshman, you probably talked about this aspect of academic writing using the term motive. Gordon Harvey of Harvard University has defined motive as [T]he intellectual context that you establish for your topic and thesis at the start of your essay, in order to suggest why someone, besides your instructor, might want to read an essay on this topic or need to hear your particular thesis argued—why your thesis isn’t just obvious to all, why other people might hold other theses (that you think are wrong)2. Put another way, the “so what?” part of your research proposal needs to explain to anyone reading your research exactly why they should spend their time doing so. Your paper’s motive tells us exactly what you are contributing, even in a small way, to a body of knowledge in the social sciences, and why that research matters in the first place. Motive is different than your actual findings – although those are important too. Motive establishes your work as part of an ongoing conversation in the academic community, and as such tends to take one of the following forms, regardless of academic discipline: 1. The truth isn’t what one would expect, or what it might appear to be on first reading. 2. The knowledge on the topic has heretofore been limited. 3. There’s a mystery or puzzle or question here that needs answering. 4. Published views of the matter conflict. 5. We can learn about a larger phenomenon by studying this smaller one. 6. This seemingly tangential or insignificant matter is actually important or interesting. 7. There’s an inconsistency, contradiction, or tension here that needs explaining. 8. The standard opinion(s) need challenging or qualifying.3 All good academics wrestle with the question of motive in their writing. Consider some of the following examples: From Douglas S. Massey, Andrew b. Gross, and Kumiko Shibuya. (1994). “Migration, Segregation, and the Geographic Concentration of Poverty.” American Sociological Review 59: 425-­‐445. Despite mounting evidence of the deleterious consequences of concentrated poverty, there is substantial disagreement about its causes. Generally, three hypotheses have been advanced . . . 2
3
See “Elements of an Academic Essay” by Gordon Harvey, available at www.princeton.edu/writing/elements.doc Kerry Walk, “Motivating Moves,” The Writing Program, Princeton Unviersity W r i t i n g S o c i o l o g y P a g e | 11 These three hypothesized causal mechanisms – net middle class out-­‐migration, net downward social mobility, and racial residential segregation – are not mutually exclusive, of course. It is quite possible, even likely, that all three operate to some extent to influence the class composition of specific neighborhoods. The relevant issues for social scientists is which hypothesis is empirically most important in accounting for the geographical concentration of black poverty, not which one is ultimately “correct.” Because relevant data are scarce, however, research on this issue has been limited, particularly with respect to class-­‐selective migration, and most of the evidence marshaled to date has been indirect (pp. 426-­‐427). Here, Massey and colleagues make at least two motivating moves: they point out that sociologists do not agree about the causes of growing concentration of poverty (motive #4), and instead generally argue for three different hypotheses, which may or may not be mutually exclusive. Second, the authors here argue that the research to date on this topic has been limited by the availability of suitable data (motive #2). From Michael Hout and Claude S. Fischer. (2002). Why More Americans Have No Religious Preference: Politics and Generations.” American Sociological Review 67: 165-­‐190. The minority of American adults who claim no religious preference doubled from 7 percent in 1991, its level for almost 20 years, to an unprecedented 14 percent in 1998. This trend is likely to surprise the many researchers who have described Americans as especially religious (e.g., Caplow 1985; Inglehart and Baker 2000), those who included religiosity as part of "American exceptionalism" (e.g., Greeley 1991; Lipset 1996), and the many observers who thought the 1990s were a time when religion was ascendant in the United States (e.g., Kohut et al. 2000). For the preference for no religion to double in less than a decade is not only a startlingly rapid social change in its own right but also a challenge to these widely held impressions of American culture. . . . We seek to explain why American adults became increasingly likely to express no religious preference as the 1990s unfolded (p. 165). In this influential article, Hout and Fischer set out to explain the puzzle (motive #3) of why more Americans list “none” as their expressed religious preference while also explaining why this seemingly tangential matter is actually important: it is “a challenge” to other “widely held impressions of American culture” (motive #8). From Stanley Lieberson and Eleanor O. Bell. (1992). “Children’s First Names: An Empirical Study of Social Taste.” American Journal of Sociology 98: 511-­‐554. Why do specific tastes take the form that they do? Obviously there is no simple answer to this question, if only because many idiosyncratic personal experiences affect individual choices. Nevertheless, in advanced industrialized societies, most expressions of taste and fashion are affected by formal organizations that have a stake in the outcome and are dedicated to influencing such tastes. . . . Sociologists in the United States rarely study first names despite their W r i t i n g S o c i o l o g y P a g e | 12 distinctive theoretical potential and despite the existence of excellent large-­‐scale data sets as well as obvious opportunities for surveys and field studies (exceptions include Miller [1927], Rossi [1965], Taylor [1974], Lieberson [1984], and Alford [1988]). Yet, the first names given to children provide a rare opportunity to study tastes in an exceptionally rigorous and systematic way that is relatively free of organized efforts to determine the outcome. Finally, Lieberson and Bell motivate their innovative study of children’s names not based on the intrinsic interest of this topic, but by presenting it as an excellent case study of the larger processes by which “specific tastes take the form that they do” (motive #5). In introducing two rival explanations as well – people’s tastes are just a result of their personal preferences, as contrasted to the idea that tastes are the result of the influence of organizations that seek to influence these preferences – the authors also suggest that a tension exists between these two rival explanations, and that using data on children’s first names can help to better explain the inner workings of tastes as they develop over time (motive #7). Motive in Student Research Looking at these three examples, it should be clear that not only does all academic work offer some statement of motive, but that really outstanding work has multiple motives (you might even argue that many of the motivating moves listed above are really just variations on the same thing). The good news here is that once you find your research question, you can continue to think about that question in dialogue with the suggested motivating moves above to think about how your independent work makes use of at least one of these motives. How do you get there? Note that in all of the examples above, the authors cite literature from their fields in situating their research question in a larger body of study and research. Research papers that are well-­‐motivated always do so in dialogue with a larger literature, situating their research in an area that others clearly care about and have worked to better understand and define. As you work to motivate your research question, remember that one of the best uses of your time is to think about how your research question can inform and be informed by the scholarly writing on your topic. How does your question help to adjudicate between competing explanations? Contribute to our knowledge? Illustrate a larger process or phenomena? A final note: remember that not only do you need to be clear about how your research matters, you must make this clear to the reader! Take nothing for granted: spell it out for your audience. Determining just who that audience is – beyond your professors and the TF– is the subject of the next section. W r i t i n g S o c i o l o g y P a g e | 13 Chapter 3: The Literature Review As you may know, researchers in all disciplines refer to the secondary sources in their discipline as “the literature.” Referring to scholarly articles and books in this manner only reinforces the importance of the written word and the cumulative efforts of scholars of all stripes to advance our shared knowledge and understanding of various subject matters. The world of “the literature” exists to inform every aspect of that collective enterprise: scholars (and book and journal editors and reviewers!) have taken untold amounts of time and resources to share with others what they’ve found in their research. Consulting this literature is absolutely essential in the early stages of your research. As you consult the literature on the topic or topics you are considering for your project, you will want to do so with specific goals and specific techniques. Below, we will consider four main questions: 1) How to find sources, 2) What you want to know, 3) How to record what you learn, and 4) Writing your literature review Finding Sources For books and journal articles, the first place you will want to start is . . . the library! And as tempted as you will be to use Google (or even Google Scholar, or Google Books), there is a reason that the Yale University library spends a significant amount of money to have access to dozens of electronic catalogues and databases: they are better than what anyone can get for free on the internet. The scholarly journal articles you will want to access for your literature review, for example, will rarely be available in full text version on the web. You can, however, access them easily (and for free!) from any computer on campus – and if you’re away from campus there are ways to access them as well via your net id and the library website (see the library website link for “Connecting from Off campus” for instructions). 1. Electronic Databases For journal articles, the best place to start for sociology will be what the library calls “core Resources” for Sociology, which can be accessed here. You can also access this page easily by visiting the Library home page -­‐> Subject Guides-­‐>Sociology-­‐>Sociology: Research Guide-­‐>click the “Core Resources” tab at the top. The Library’s Sociology Research Guide can be accessed here. Most of these databases listed in these guides will contain full text versions of sociology’s most important journals. You can pick the database with the interface you feel most comfortable with, and of course you can talk with the library’s Sociology librarian, Gwyneth Crowley, about how to find the resources you need for your topic. Her contact information is also listed in the Appendix to this handbook: (203) 432-­‐3213 or [email protected]. W r i t i n g S o c i o l o g y P a g e | 14 2. The Library Catalogue You will find books in Yale’s library by visiting http://web.library.yale.edu/. On one hand this seems pretty obvious, but it’s worth emphasizing: you will actually have to go and find hard copies of books in the stacks for your senior thesis! Google books and ebrary do contain some titles, but most of the titles digitized in these electronic formats are skewed toward older works in the humanities. Social science research – most of which is concerned with current phenomena, and thus are likely to still be in print – are usually only included sporadically and not in full text form. Getting the hard copy is also useful for additional research, as discussed below. 3. “Shoe-­‐leather Detective Work” A final strategy that you may find useful for finding sources could be called simply “Shoe-­‐leather” detective work, and presents another reason why finding hard copies of books is essential for your research. Look in any academic volume and you will find some section of Acknowledgements in which the author thanks the colleagues who have read all or portions of her work. Remember that academic work is a collaborative enterprise, so these people mentioned in a book’s acknowledgements will not simply be the author’s friends and buddies (although they may well be these, too), they will be other scholars whose research addresses similar topics and issues. You can also find paperback copies of books that will have blurbs by other scholars praising the author’s work: find these people in the library catalogue, and see if their work also informs your topic. Likewise, look at the bibliography to books that you find especially useful for your topic – these will, of course, be relevant for your own research. You can use this same strategy by simply looking up citations in journal articles that you find concerning your topic. A great resource for this are the articles published in Annual Reviews of Sociology (www.annualreviews.org). The articles prepared for this annual volume are intended to exhaustively summarize the state of a particular field of research in sociology, and your first stop on the road to conducting a thorough literature review should be to look up your topic and find related articles in this publication. The citations listed as works cited will be important to cull through as well. What You Want to Know When you go to this literature, you must have some principle goals in mind or else you will find yourself overwhelmed with information. Your first impulse might be to try to summarize every source that you read in condensed form – their use of other sources, data, findings, and so on. A better way to use your time is to think first about how you need to use these sources, and particularly about how you might use them in better conceptualizing your research question. Keep in mind that scholars cite sources at a variety of levels, ranging from larger, big-­‐picture points to smaller, more detail-­‐oriented information. In a well-­‐written literature review, you will probably want to W r i t i n g S o c i o l o g y P a g e | 15 do some combination of these things. Put another way, you will probably end up citing some sources as “forests” and others as “trees” – the forests offer broad, general points while the trees focus on more specific information. For example, the book Unanticipated Gains: Origins of Network Inequality in Everyday Life by Mario Small (Oxford University Press, 2009) could be cited in any of the following ways: 1. Very Broad: Distills the research to its most general, central point Recent research by Small (2009) indicates that individuals’ social networks are shaped in important ways by the organizations they interact with on a daily basis. 2. Mid-­‐Level: Describes some of the book’s more specific findings from its focus case study Small (2009) shows that child care centers help create network ties for the mothers who use them – for example, by offering referrals to other organizations and helping mothers form relationships with other center patrons. 3. Very Specific: Cites specific, technical information or statistical findings from the book’s investigations In his recent book, Small finds that almost 90% of government-­‐funded child care centers have a formal parents’ organization (2009: 81). Each of these references to Small’s work could build on this source to make a larger argument, but not all will be useful in every case. Of course, you won’t know how you might use sources until you are further along in your research process, so it’s a good idea to take strategic notes in your reading that you can refer to later when you actually write your literature review. You will therefore want to take careful notes along the way. You should take care to take a few notes about every source you consult that describe its broad contribution, and at least a few mid-­‐level notes about the sources that are most closely related to your research topic. Very specific findings (as in the third example) will likely be the least useful, so you will probably not need to take such detailed notes in your early review of the literature. As your research question becomes more focused, you can always return to your sources for this kind of very specific data, if it is needed. Some other strategies that can be useful in your note-­‐taking are to think about the following three questions as you read and record what you find in your literature review: W r i t i n g S o c i o l o g y P a g e | 16 1. What are they asking? At the very beginning of your research, you will want to consult your sources with the goal of finding out what questions other scholars are asking about your topic. If you’re interested in the social organization of poor neighborhoods, for example, you will likely find that researchers interested in this topic ask a range of questions: how are poor neighborhoods affected by deinstitutionalization? Are poor neighborhoods really devoid of organizations and institutions? What is the role of social capital (and its decline) in racially segregated areas? What kinds of social organization emerge to fill to gaps left by formal institutions? You will begin to get a sense of a more focused research question as you get deeper into the literature on your topic and see the different pieces (recall the discussion of developing a research question in Chapter 1) of the topic and how they are related. 2. What are they finding? Since social science is about the cumulative process of understanding and theorizing the social world, a second goal in your literature review will be to note patterns in the literature about your topic. Remember that finding places where the published literature disagrees – or even contradicts itself – is a great place to situate your research question (and your motive!) Returning again to the subject of poor neighborhoods, you will probably find in your literature review that most scholars of urban poverty believe that neighborhood factors mediate the influence of poverty to some extent, but disagree (or have yet to reach consensus) about the relative importance of the exact mechanisms – for instance, what matters most for poor neighborhoods: disorganized family structures, poverty, poor schools, deinstitutionalization, incarceration? Or how do these different factors interact to create cumulative disadvantage for poor populations? 3. How do they know? Finally, in your literature review, keep close tabs on how the scholars studying your topic reach the conclusions that they do. Do they draw conclusions based on large-­‐scale survey data? On interviews with a small sample of respondents? Ethnographic observations? A combination of these methods? Likewise, give some thought to ways in which different methods can supplement findings from existing studies. Keeping Track of What You Read and Avoiding Plagiarism The final task in your literature review – before you write it up, that is – is keeping track of what you read. Ultimately, how you do this is a personal preference, but it is essential that you develop some kind of organized note-­‐taking system as you undertake your research. You may prefer to take notes in a Word document, to keep track of notes in a citation management software program, such as Zotero, RefWorks or Endnote (SOCY 493 will use Zotero), or to take handwritten notes in a centralized location, such as a spiral notebook devoted entirely to your research. W r i t i n g S o c i o l o g y P a g e | 17 Whatever your strategy, you will find in the course of writing your thesis that you can’t rely on scrawling a few notes on a hardcopy printout of an article. In order to develop your thinking sufficiently to write an adequate literature review you will need to take notes in an organized way. This will save you time later – when you go to write up your literature review (see the final section in this chapter), you will find that much of the work is already done. If you don’t take notes while you’re reading, you’ll find you have to do much of the work over again when it comes time to write the review. In fact, you can get a head start on this part of the process by actually writing about what you’re reading in your review of relevant literature. You accomplish two things by doing this: one, you take notes on the source you’re reviewing, and two, you clarify and develop your thinking about that source by being in conversation with it through your own writing. This will also help your independent work because writing is a critical part of the research process. In fact, it’s a mistake to think of research and writing as being two separate processes: people who write every day tend to be more productive in their research and, not surprisingly, their writing as well. Aside from helping you to be more productive and efficient, taking careful notes on what you’ve read is important for another, more serious reason: it helps protect you against plagiarism. At this point in your academic career, you know well what the definition of plagiarism is (the use of a source without proper acknowledgement) and its consequences at the university (severe). As you begin working on your thesis, it is absolutely critical that you distinguish your ideas from the work of others, and careful notetaking is a vital part of this process. You may find it helpful to paraphrase as you take notes in order to avoid accidentally citing verbatim from a source in a later document. Or you may prefer to take notes by selecting a few key quotes from a source (and noting them as such) and then writing about them in your own words. Whatever strategy you chose, careful note taking is essential. You might not set out to plagiarize in your independent work, but could do so inadvertently through sloppiness, fatigue, or sheer laziness in your study habits. Keep in mind that the University does not take intent or motive into account when evaluating allegations of plagiarism. A complete outline of the University’s policies can be found in the Yale Undergraduate Regulations Handbook. Using Sources and Writing Your Literature Review Now that you’re well along the way to reviewing the relevant literature on your topic, you will want to organize your literature review in a way appropriate to your particular topic and research question. Since you will probably feel as though you’ve become an expert on your topic, you may feel tempted to display every single fruit of your labors by regurgitating every single piece of information you gleaned in the course of weeks of library research. You should resist this temptation! This is one of the sad truths of scholarship: much of what we do does not every see the light of day. You may have reviewed two dozen articles before you found four that W r i t i n g S o c i o l o g y P a g e | 18 really informed your research question. Perhaps you changed or adjusted your question midway through your literature review as you found that much of what you’d been reading was irrelevant, or concerned a topic that was already well-­‐studied. It is only natural that you would want to get some credit for all of this hard work. But even if it is ultimately not useful to cite or discuss all of these sources, the time you put into reading them was still essential because you could not have arrived at a worthwhile research question without surveying the literature in the field to discern where your contribution could be most valuable. The way to show the fruits of your labor is to write a well organized and pointed literature review that explains just how your research question fits into the established literature in its field in sociology, and promises the reader that your research will contribute something worthwhile. In organizing your literature review, it’s a good idea to revisit the motivating moves discussed in Chapter 2, and to give some thought to how your research speaks to at least one (or more) of these motives. Is there a conflict in the literature that your research can address? A gaping hole in existing scholarship that your work will fill? Or perhaps your thesis will argue that we can understand a larger phenomenon through your analysis of a smaller case study. Whatever the contribution, your literature review should make clear the state of the field on this topic and exactly how your research will improve it. As you draft your literature review, you might find it helpful to use some of these questions to help and organize your notes into larger themes, or as a jumping off point for organizing your larger literature review: • What are the main debates in the field concerning my topic? • What are the key ways in which scholarship on my topic has evolved in the past several years (or decades)? • What hypotheses might the established literature suggest for my particular research question? • How have other scholars attempted to answer my research question? In what ways is our knowledge still incomplete? • What are the most important findings about my topic or question? Unfortunately, there is no clear-­‐cut formula to writing a good literature review, but all good literature reviews share the common feature of reviewing literature for a purpose, which is to motivate the investigation of a new research question. This is why you will want to resist the urge to summarize and report on all that you’ve found – your task in writing about your independent research is to use the sources you’ve found to place your research question in a larger context of significance. In effect, your literature review creates an academic home – and thus, an audience – for your research. W r i t i n g S o c i o l o g y P a g e | 19 Chapter 4: Finding Data As you may know already, data in the social sciences is divided into two broad categories: qualitative and quantitative data. Put simply, qualitative data consists of data measured primarily in words, while quantitative data measures data as numerical values. If you will be using qualitative methods in your thesis, you will most likely need to collect your own data – be it through observation, interviewing a relevant sample, or analyzing text materials through the methods of content analysis. Depending on the nature of your research, you may need to file an application with the university’s IRB committee for approval of research using human subjects. We discuss that process further below. The methods used by sociologists to analyze quantitative data generally involve statistical analyses, through the use of computer software such as STATA or SPSS. If you want to use statistical methods in your independent research, you need to have some training in these methods. You may need to take an additional course this year if you choose to use statistical methods. You will also need to find a dataset suitable for analyzing your research question. Although this may sound difficult, you will find that there are innumerable datasets made available to you at Yale, and many more that are available in the public domain. Depending on your interest, faculty at Yale may also have access to data collected through their own research that you may receive permission to use for your independent research. This chapter proceeds by introducing the IRB process (typically required for qualitative studies) and then describing places that you can locate existing data for your research (if you will be using quantitative methods). The Institutional Review Board (IRB) All universities have an internal body that is responsible for making sure that research on human subjects is ethical and takes adequate care to protect the confidentiality of the subjects involved in the research. This means that you will need to get approval from Yale’s Institutional Review Board in order to conduct research that involves interviewing human subjects in the form of personal interviews or survey responses administered by you. You can get more information by visiting the university’s human subjects protections website. However, a quick summary of the IRB policy is reprinted here for your convenience: Proposed research including journalistic interviews, oral histories, biographical profiles, or other forms of nonfiction narratives, normally does not fall within the jurisdiction of the IRB. In these cases, the individuals being interviewed understand that they are being quoted, and have every expectation that their views will be made known. The interviewees are advised of their right to remain anonymous, to have their remarks printed without attribution, or kept ‘off the record’. If the interviewee is directly quoted, they are allowed to read or hear the quotations attributed to them. The interviewee will also be advised of any publication plans for the project. Most projects from Humanities meet the above criteria; therefore they do not qualify for the IRB review, and do not need to submit their project to the IRB for approval. W r i t i n g S o c i o l o g y P a g e | 20 Projects that can be defined as a systematic investigation which includes research development, testing and evaluation, designed to contribute to generalizable knowledge, including; surveys, questionnaire studies, or examinations of existing data, do fall within the jurisdiction of the IRB. To put this into plain English, let’s consider a hypothetical example. Suppose you are interested in the ways in which high school teachers interact with at-­‐risk students. You might want to interview teachers or school principals on this topic, and could do so in two ways: The first way would involve interviewing superintendants of school districts (public figures) as informants, which would mean essentially approaching them in the same way that journalists would their subjects: you want to speak with them in order to get information about a particular problem, and the subjects understand from the start that you are using them for this purpose and that you will quote them in your research. You will plan to get back in touch with them before you make your findings public in order to confirm that you have quoted them correctly, and advise them that they don’t have to be quoted by name if they do not wish to. Although the person is being interviewed, the subject is an aspect of the educational process – at risk students and their experience in the school system. This form of research does not need IRB approval. But you might also pursue this research question and your interviews in a slightly different way. You might really be interested in teachers’ experience of this interaction, and wish to talk with educators about their own experiences with at-­‐risk students and how they have personally handled situations with this population, as well as their own feelings about those interactions. In this case, your research subject is slightly different – the activities and experiences of the individuals themselves. If you are interviewing teachers in this way (the subject being their own person and personal experience), with the goal of making inferences about a larger population of research subjects (the larger population of teachers of at-­‐risk students), then this research must be approved by the IRB and you will need to take steps to preserve their confidentiality and secure their consent prior to the research. Why the difference? In the first example you are using human subjects to get background information about a specific topic, in the second example case you are gathering information about these human subjects themselves. In that case, the IRB must approve your research. Students often ask questions about whether or not they need to file an application with the IRB in order to do ethnographic research, or participant/non-­‐participant observation in a public setting. The answer is generally yes – you do need to file an application if you are going to be observing human behavior and making inferences based on those observations. You will not need to gain participant consent in this setting – if you don’t actually talk with people and ask them questions, they won’t know that they are actually the subjects of research. However, this makes the IRB process even more important: you must be prepared to describe the goals and benefits of your research, and the steps you will take to ensure the privacy of your research subjects in this setting. And if you plan to do a combination of field observations and interviews, then you will definitely need to file an IRB application. W r i t i n g S o c i o l o g y P a g e | 21 Is my research exempt? As if the IRB process were not already significantly confusing, even when you realize that you must file for approval with the IRB you may not be sure whether or not your project is exempt. In general, these exemptions cover certain kinds of research in educational settings, and research with human subjects (including interviews and surveys) that will cover “non-­‐sensitive” material, with anonymity of the respondents guaranteed. For example, if you plan to survey people randomly, without recording their names, this sort of research is likely exempt for IRB purposes. If you are going to interview respondents and keep a record of their real names, or collect information about them that might potentially render them identifiable and thus jeopardize their confidentiality, then it is not. If you aren’t sure, you should talk with your professor and the TF for advice – and when in doubt, it is probably safer to file a long form of the application. If you file your project as exempt and the committee decides your research is not exempt, you will lose time and have to submit a new form and wait for the committee’s next convening. The IRB meets regularly; consult their website for a schedule of meetings. You will need to stay on top of these dates in order to submit your application soon enough to give you adequate time to collect your data and analyze it for the various due dates that come up in the spring semester. Your IRB application will also require the signature of an advisor, and require you to say a good bit about your research project and prepare relevant documentation for your research, such as a proposed interview questionnaire and consent form for your respondents. In other words: don’t wait until the last minute! You should plan to give your advisor all of these materials when you ask for his or her signature on the application – allowing two weeks before the IRB’s submittal deadline for your advisor to review your application is a good idea. On final point: regardless of any final IRB application, students enrolled in SOCY 493 do need to complete the IRB training certification online, by visiting https://bmsweb.med.yale.edu/tms/tms_enrollments.offerings?p_crs_id=160&p_std_id= . Check your syllabus for when this is due in class; typically you will need to submit your certificate to your teaching fellow by the week after fall break. Working With Quantitative Data Yale has a number of resources designed to help students using quantitative data, starting with the Center for Science and Social Science Information (CSSSI), located in the Kline Biology Tower. This office employs people whose entire job concerns helping members of the Yale University community find and analyze quantitative data. They are there to help you, and you should make full use of their services if you want to write a quantitative senior thesis. You can get a good idea of the wide range of data that are available to you as a Yale student by visiting this website. This web page offers access to collections of data that have been collected by other researchers, and which are now available to the public (or at least, the public affiliated with institutions of higher learning) at no charge to you, as a student. If you choose to analyze one of these data sources, you will need to do so in its original form – in other words, you won’t be able to make claims based on W r i t i n g S o c i o l o g y P a g e | 22 this data as it is reported in another scholarly article or book; you will need to download it and analyze it yourself in a computer program such as STATA or SPSS. Courses like SOC 493 will certainly help to prepare you to do this; you can also turn to the CSSSI librarians and subject specialists for help. They have drop in hours which are listed on http://guides.library.yale.edu/friendly.php?s=data-­‐statistics. See also the Appendix to this guide for a list of important contacts for help finding data in different subject areas such as economics and finance, politics, and public policy. Other Sources of Quantitative Data As a Yale student you may also be able to conduct research on data that is not yet publicly available – this would mean consulting a faculty member about projects currently underway and receiving permission to explore certain questions using data from their projects. To get a better idea of what Yale faculty are currently up to, visit http://sociology.yale.edu/. If you see a project described on this website that sounds like it might fit your interests, by all means email the faculty member(s) associated with it and ask if you could meet with them to discuss their project and the possibility of working on a small part of the data for your independent work. Depending on limitations of time and the structure of the particular projects, this may not always be possible – but it is completely appropriate that you ask to discuss the opportunity. The worst that can happen is that the professor will politely decline due to time constraints or the limitations imposed on the project by its funders or other collaborators. The best that could happen is that your senior thesis could analyze data not currently available to the wider scholarly community! But you won’t know unless you ask. W r i t i n g S o c i o l o g y P a g e | 23 Chapter 5: Methods Sociologists use a variety of methods in their research. As a student, all of these methods are available to you as you conduct the original research required by your senior thesis. The methods you use in your independent work will depend on three things: the research question you select, your own individual preferences, and the constraints of time and resources. First, the research question you select will largely determine the methods you will use in your independent work – or at the very least, make some methods less appropriate and others more so. As discussed in the previous chapter, the main methodological distinction in the social sciences comes between quantitative and qualitative methods. If you use quantitative analysis you will almost certainly be using the methods of statistics to analyze a publicly-­‐available dataset. Conversely, qualitative data analysis includes methods such as content analysis, interviews, and ethnographic research. Sociologists also use experimental methods, which may be either qualitative or quantitative in terms of the data collected and subsequently analyzed. Below we discuss each of these methods in brief, along with some elements of each methodology that require particular attention from you, the researcher. Survey Data Quantitative methods in sociology typically involve the analysis of survey data. Some students working on senior theses use existing data sets (see the suggestions included in the preceding chapter) while others will devise their own surveys using any number of existing web-­‐based applications. The first approach – using existing data – is a bit more straightforward and efficient, but you may have to do some work to find a data source the includes the questions or topics you’re interested in studying, but that hasn’t already been analyzed in exactly the same way in the existing literature. On the other hand, constructing your own survey will allow you to ask the exact questions you’re interested in, although this approach typically requires more time and investment at the start of the research process. If you devise your own survey, you will need to file for IRB approval in a timely manner (see Chapter 4) and should plan on devoting intense time and effort into crafting your survey questions and design before you initiate your research. It is also advisable to run a pilot of your survey to identify points of confusion or potentially problematic questions on your survey instrument. Also, keep in mind that in order for your analysis of a survey to be feasible, the dependent variables you’re interested in need to have substantial variation in value. In other words, your respondents need to give different answers to the survey items you’re interested in analyzing, or else there won’t be anything to study. If you decide to construct your own survey, you will also have to decide how you will distribute it to potential respondents. In some circumstances, students can apply for departmental funding to distribute your survey to a random selection of respondents through a data collection company. This will give you a random selection of respondents (allowing you to potentially make generalizable inferences based on your data), although the constraints of resources will likely mean you have a smaller end number of respondents. You can also distribute your survey through friends and family through an email web link, which may result in a larger number of respondents. However, you will not have a W r i t i n g S o c i o l o g y P a g e | 24 random sample, and your results and analysis may be skewed as a result. This is not an insurmountable problem, but one that you will need to be prepared to take into consideration as you write about and analyze your data. Finally, using quantitative methods requires that you have taken (or are taking) the appropriate courses in statistical analysis, and the use of the appropriate computer software. STATA and SPSS are available on most cluster computers at Yale, and student versions of the software are available for purchase at an educational discount. Keep in mind that particular kinds of data demand particular kinds of analysis. For instance, it’s likely that the courses you will have taken in statistics conclude with Ordinary Least Squares (OLS) regression, which is a statistical methodology appropriate for analyzing a continuous outcome. This means that if you construct your own survey, for instance, it’s advisable to construct your outcome measures as a continuous variable (e.g. 1-­‐5), because OLS is not an appropriate form of analysis for variables with a yes/no (1/0) outcome. Students should feel free to learn and implement advanced statistical techniques. Interviews Interviews collect data from respondents based on an interview guide that asks similar questions of everyone. You’ll need to invest some time in developing your list of questions, and you should decide if your interviews will be open ended or more structured (in the latter case, it is usually advisable to put the questions in a particular order). While you may not conduct every interview in exactly the same manner, sociologists using structured interviews generally strive to ask questions in the same way of each respondent, placing questions that are more delicate or require more trust at the end of the interview, when rapport has been established. You should also think about ways to avoid biasing your respondents based on your description of your research, or early questions in the interview. Finally, interview methods require that you submit the appropriate IRB application, including a consent form that your respondents will need to read and sign to indicate their willingness to participate in your research. If you choose to use interview methods, you will also need to decide how you will find a population to interview. In most cases, how you find your respondents will determine who they are, as well as the data you gather. For instance, if you want to interview mothers of young children, you could post a flyer (with permission, of course) at a local day care center, or you could hang out at the Cotsen children’s library at the New Haven public library. However, these two populations of mothers may vary in important ways: the former will likely be employed, the latter fulltime caregivers. In either case, using interviews requires a good bit of attention to logistics, as it requires that you contact respondents and arrange to interview them at a specific time. Interviews will typically last 1-­‐2 hours. Interviews are a particularly useful way of gathering information about people’s interpretations and motivations – why people do things, and what different experiences and social situations mean to them. In the above example, interviews would be an ideal way to ask mothers how they feel about a range of things that have to do with the care and development of children. Because they are not drawn from a W r i t i n g S o c i o l o g y P a g e | 25 representative sample, however, interviews are less well-­‐suited to gathering information about the distribution of certain characteristics among a population – say, the number of children per household. Ethnography Ethnography involves sustained observation of people in a particular social context. Some ethnography will involve participant observation (in which you engage in the activity or group you are studying) while other ethnography will be strictly observational. In most cases, however, ethnography requires IRB approval, especially because researchers often use the observational methods of ethnography in conjunction with interviews. The strength of ethnography lies in the depth of interpretation that this method can bring to everyday phenomena – what it means to hang out at the library, go to a sporting event, be a member of an eating club, and so on. Although Clifford Geertz famously called ethnography “deep hanging out,” ethnography requires much more than that, since the researcher must take detailed field notes that record the researcher’s observations. Taking field notes requires that you make brief notes while in the research setting – often called “jottings” – to help you remember key events and observations from your field setting. If you spend several hours observing shoppers in a grocery store, for instance, you might keep handy a notebook to write notes about specific instances that you observed. Later (preferably the same day, or at most the next day) you should revisit your jottings to write down extended field notes. Writing field notes is a laborious process, but this is an essential step in qualitative field work. If you wait (or worse, neglect) to write field notes you will find that you have little data to return to for analysis in your independent research. Content Analysis Sociologists also study text, such as newspaper articles or websites. Content analysis looks closely at the ways in which ideas, topics, or information are communicated in textual data. For instance, you might decide that you want to study the recent debate over health care reform, analyzing the ways in which arguments for and against “Obamacare” were communicated in the editorial pages of key national newspapers. Content analysis typically requires that the researcher make a number of key decisions about how to gather, code, and analyze data. In this case, for example you would need to decide which newspapers to study, and why. Do you want to study the newspapers with the highest numbers of national circulation, or regional newspapers in different parts of the country (or both?) What time period will you study? And finally, what will you code in the articles, and how will you decide the unit of analysis (e.g. article, claim, word)? For these reasons, content analysis demands that the researcher spend a good bit of time at the beginning of the research coding sample documents to look for larger patterns or commonalities in the data. In content analysis, you won’t be able to code everything about the text, but with some time and thought you will be able to decide what about your data is most interesting for the research questions you’ve chosen to pursue. Although it’s not essential, many students and veteran researchers alike find it useful to use a qualitative textual analysis program like Atlas.TI or NVivo to keep track of your data and observations. W r i t i n g S o c i o l o g y P a g e | 26 Finally, although content analysis makes use of (qualitative) textual data, some content analysis will also make use of quantitative comparisons. Particularly if you code a larger number of documents, you may find it useful to communicate some features of those documents in numerical form. For instance, in the above example it might be useful to report the percentage of editorials that made claims supportive of Obama’s health care reform plan in different newspapers. Experimental Designs While experimental methods remain less common that other methodologies in sociology, these methods can be ideal for investigating certain phenomena. The main features of an experiment involve comparing key features of two groups: one that we call the “control” group, and a “treatment” group in which one key variable of the study design has been altered. For instance, you might be interested in studying the stigma surrounding obesity and decide to construct a vignette about an individual suffering various hardships. You could have respondents read one of two versions of the vignette (either in an interview setting, or on a survey): in one version you mention that the hypothetical protagonist is overweight, while the other version omits this detail. You could ask respondents a series of questions about the vignette and how they think about the protagonist’s problem, and compare the results among the “treatment” group who were told that the individual described in the story was overweight, as compared to the “control” group. An experimental design also requires that individuals be randomly assigned to either the treatment or the control group, and that the entire population under study be comparable. In other words, if you use experimental methods in your research you should take care to make sure that the two groups are comparable to one another (in other words, your research subjects assigned to each group need to be drawn from the same population). Experimental methods will require that you file an IRB application. Mixed Method Designs Some senior theses might use “mixed methods” research, meaning that the researcher uses some combination of the quantitative and qualitative methods described above. The use of more than one methodological approach is increasingly valued in sociology, and many researchers agree that the use of both survey research and interviews or ethnography give a more complete picture than the use of one method alone. Writing About Methods Yet in spite of all the attention paid to research methods in the social sciences, sociologists rarely talk much about the process of writing about methods – and this step is critical in the process of writing about your research. This remainder of this chapter is designed to help you improve the ways in which you write about your methods in your senior thesis. To some extent, how sociologists write about their methods is common across all forms of methodological inquiry. Regardless of their means of analysis, sociologists must do the following things: 1. Describe your sample of data, and explain how it was constructed W r i t i n g S o c i o l o g y P a g e | 27 2. Define variables, means of measurement, and any hypotheses you have 3. Describe the methods by which you will analyze this data How you do these three things will depend somewhat on your methodological approach. Below we will look at examples of each task from the published literature. 1. Describing a sample At a minimum, you must tell your reader exactly what data you are analyzing, and how you arrived at that data. If you are using a quantitative data set, you must tell us what the survey is, when it was conducted, and how many respondents it includes. If you are engaging in qualitative research, you must tell us how you gathered your data – if you interviewed a group of individuals, you must tell us how you found them (Did you ask for referrals from friends? Approach strangers on the street?) If you conducted ethnographic fieldwork for your pilot study you will need to describe your research setting to the reader. Accordingly, your readers will want (and deserve) to know how your sampling procedures may affect your findings – if you interviewed Yale students about their tastes in music, for example, you will need to explain in your proposal why Yale students are worthy of investigation, and how they may be different (or perhaps atypical) as compared to other college students. Here are some examples of how sociologists describe their sample in the published literature: Example 1: A Qualitative Sample From Ali M. Ahmed. (2009) “Are Religious People More Prosocial? A Quasi-­‐Experimental Study with Madrasah Pupils in a Rural Community in India” Journal for the Scientific Study of Religion 48: 368-­‐
374: A total of 102 men participated in the experiment, which was conducted in March 2006 in a small town called Nowganwan in Uttar Pradesh in India. Forty-­‐two were students in religion and theology, imams-­‐in-­‐training, recruited from two madrasah's (religious Muslim schools) and 60 were students in the social sciences recruited from the local men's college. (For the sake of simplicity I will from now on refer to the imam students as religious students and to the other students as nonreligious students.) The average age of both the religious and nonreligious students was 17 years. The students were informed about the study in regular classes. Those who wished to participate in the study signed up on a list that was circulated in the class (p. 369). Here the author tells us several useful things: that the study compares two groups of students (religious and non-­‐religious), the number included in each group, and where and how the subjects were recruited. The author also tells us that the groups are comparable in some important ways: they are of roughly the same age, and were identified through identical sampling methods (the study was publicized in educational settings, encouraging voluntary participation.) Example 2: A Quantitative Sample W r i t i n g S o c i o l o g y P a g e | 28 From Richard J. Petts (2009). “Family and Religious characteristics’ influence on Delinquency Trajectories from Adolescence to Young Adulthood.” American Sociological Review 74: 465-­‐483: The data for this study are taken from the child and Young Adult Sample of the 1979 National Longitudinal Survey of Youth (NLSY79). Original respondents of the NLSY79 were between the ages of 14 and 21 years old in 1979, and Black and Hispanic respondents were oversampled. For the child and young adult sample, information was collected from each child of the female survey respondents age 10 years and older biennially since 1988. In addition, a young adult survey was administered to children age 15 year and older biennially since 1994. To construct the sample for this study, I first pool data from 1988 through 2004 according to the youths’ age. I then restrict the sample to youths interviewed at least once in early adolescence (ages 10 to 14), resulting in a sample size of 6,693 (p. 470). This second example is very detailed, but tells us some important things about the author’s sample. First, the author explains that this is a smaller subset of a large, well-­‐known data set (the NLSY79) but that the author is analyzing only the responses of the children of female respondents , and only those children who were interviewed at least one time when between the ages of 10 and 14 at the time of the survey. Regardless of whether you are writing about a sample of individuals that you have collected (as in the first example) or a large, publically available dataset (as in the second), you must describe your sample by telling your readers how many people are included in your sample (or articles or other sources of evidence in content analysis), how they were selected, and when the data were collected. In qualitative analyses, you generally need to say a bit more about how people were recruited for the study. When appropriate, you should do some theorizing about how your methods of recruitment might influence your conclusions – in effect, you must address the issue of how generalizable your findings are. 2. Define Variables, Means of Measurement, and Hypotheses If you are writing about quantitative data that you analyze statistically, the task of defining your variables is a bit more straightforward, as you must make these decisions clearly at the beginning of your analysis. Consider the following examples. Example 1: A Quantitative Analysis From David M. Harding, (2009). “Violence, Older Peers, and the Socialization of Adolescent Boys in Disadvantaged Neighborhoods.” American Sociological Review 74: 445-­‐464. For older peers to play a role in the differential socialization of adolescents in poor neighborhoods, such adolescents must experience greater interaction with them. I investigate whether this is the case using nationally representative data from the National Educational Longitudinal Survey (NELS). In the first follow-­‐up, conducted two years after 8th grade respondents were sampled, adolescent were asked about the ages of the “people with whom W r i t i n g S o c i o l o g y P a g e | 29 you spend most of your time.” I use responses to this question in conjunction with the respondent’s age to construct an indicator for whether the respondent spends most of his time with individuals who are older than him but under age 26. Here, Harding explains how he is constructing a key variable in his analysis: whether or not respondents report that they spend most of their time with older peers. He explains that he uses respondents’ answer to this question (the age of their main companions) along with the respondent’s age to determine whether or not the respondent spend a considerable amount of time with older individuals. In qualitative research, defining variables can be a bit more complex, but is no less important. The key to describing measurement in qualitative research is to make clear to the reader the kinds of decisions you likely had to make early on in your analysis about what to code in your data, and about how to code it. Consider the following example, from a qualitative analysis of the Turkish sermons: Example 2: Content Analysis From James Gibbon (2008). “God is great, God is good: Teaching god concepts in Turkish Islamic sermons”. Poetics 36: 389-­‐403 Sermons were coded using TAMS Analyzer, a Macintosh application similar to ATLAS.ti. Codes were applied at the sentence level to all text that appeared in the body of the sermons, including scripture. […] Attribute codes were assigned when a sentence contained a declarative statement about God or implied an attribute by describing an action performed by God (e.g., the sentence ‘‘God created the world’’ was coded CREATOR). Actions were coded as attributes only in cases where the action exceeded normal human capability or corresponded to a trait mentioned elsewhere in a declarative sentence. This rule prevented me from having to decide arbitrarily whether prosaic sentences like ‘‘God stated. . .’’ or ‘‘God considers. . .’’ revealed character traits or not (p. 394). Here, the writer explains several things: the level of analysis at which he applied the codes (the sentence level), and how he made key decisions, such as a prayer’s designation of certain character traits to God – likewise, he how he dealt with ambiguous cases. Regardless of your methodological approach, if you have any specific hypotheses you should state them somewhere in this discussion. Note that a hypothesis does not always need to be a formal statement, but can also take the form of a statement of what you expect to find, given what you know from your literature review and your general expectations about your data. 3. Describe Your Methods of Analysis This is an instance in which quantitative analysis is a bit more straightforward: if you are analyzing the relationship between religious affiliation and years of education, for example, you will simply tell W r i t i n g S o c i o l o g y P a g e | 30 your reader that you analyze this relationship with a certain form of statistical modeling, such as Ordinary Least Squares (OLS) Regression. Explaining your methods of analysis in qualitative work can be a bit more complicated, largely because these analyses often proceed more inductively, seeking to draw larger conclusions based on a smaller subset of data. In these instances, what is most important is that you explain to your reader how you proceed with your investigation, and the how this is an appropriate strategy given your research question and its larger significance. Ann Swidler does this very well in the Methodological Appendix of her influential book Talk of Love: How Culture Matters (University of Chicago Press, 2001): I was less interested in what people thought than in what resources they had available to think with, and how they mobilized those resources. Thus the interview was open-­‐ended, both in the sense that the questions were unstructured and that I probed what people told me, following up on what they said, asking about its meaning, and trying to determine its ramifications for other things they thought. Thus, while I always covered the basic questions in the interview, often more than once, I also pursued topics of interest to the interviewee, trying to find out how she or he really mobilized her ideas to address different kinds of issues. I was especially interested in the ways cultural traditions penetrate and shape experience. Thus I asked people to describe their experience and considered how they organized that experience and what kinds of cultural meanings guided it. To understand more deeply how people used ideas, I often responded to statements of general principles (“Honesty is the most important thing in a marriage”) by asking for examples, to clarify what grounded the principle’s meaning. When people talked in specifics, I often asked for a generalization, to see how they might extend the construction they made of the concrete case (p. 221) Swidler goes on to include the interview guide used in her data collection, but some of the most important aspects of her method are summarized here: since she is interested in how people may use culture strategically (and even in contradictory ways), her methods of analysis involve asking respondents to think deeply about their ideas, probing them to come to think about the relationship between their specific statements and larger, general principles. She makes clear that what she’s interested in is not necessarily what people say but the range of resources that they draw upon to say it. A Final Point While the manner in which you write about your methods will vary somewhat according to your methods, your topic, and your specific method of analysis, it’s a good idea to err on the side of reporting too much information about your research than reporting too little. You can also include key methodological information – such as your interview guide, or descriptive statistics of respondents – in W r i t i n g S o c i o l o g y P a g e | 31 the appendix to your paper if you fear that this information will eat up too much of the body of your final draft (see the discussion of “The Final Product” in Chapter 9). W r i t i n g S o c i o l o g y P a g e | 32 Chapter 6: Data Analysis In writing about the results of your data analysis, the logistics of how you do so will depend somewhat on whether you are using qualitative or quantitative methods. But regardless of your methodological approach, your data analysis in your senior thesis will do the following things: cite evidence, discuss the interpretation of that evidence, and make some form of a larger claim based on that interpretation of evidence. The nature of the evidence you use in your independent research will depend on the methods you use in gathering and analyzing your data. Regardless of your methods, however, the early phases of data analysis are typically murky and somewhat uncertain simply because the things we attempt to study as sociologists often turn out to be more complicated, surprising or unexpected than we first thought! Your project likely started with a clear research question, but having collected your data you may find yourself entering a new phase of uncertainty. For instance: your senior thesis may have set out to examine how stay-­‐at-­‐home parents think about gender roles in a setting where a marriage is, objectively, somewhat unequal in that one partner earns all of the household’s income. Yet your interviews discovered that most of your respondents don’t think or talk very much about male or female identities when they discuss domestic work, so you’re at a loss to do next. Things like this happen with regularity in social research, and far from being a problem or an obstacle, are actually an opportunity to deepen your analysis and contributions to the field. This chapter offers some useful suggestions for how you can wade through the early phases of uncertainty surrounding data analysis to find out the most compelling features of your data. Tips for Early Data Collection: Keeping Notes Whether you are working with qualitative or quantitative data, you should begin keeping a data log or journal as soon as you begin working with your data. If you’re using quantitative data and analyzing a dataset in STATA, for instance, you should keep some form of a log in which you take three kinds of notes on the stages of your analysis. The first part of these notes should be recording descriptive information about what you did and what you found: How did you code certain groups? What regression analyses did you perform? What were the results? You’ll have records of these analyses in the form of log and do files, of course, but you will want to have additional notes to keep track of how your analysis is progressing. In addition, you should keep notes about additional questions that your findings raise. Some of these may not be answerable with your data, but others will help you figure out how to take the next step with your analysis. You will want to explore these in future stages of your analysis, so you should also write down the next steps suggested by your findings. For instance, the following might be notes produced by a project analyzing the relationship between adolescent religious tradition and educational attainment (in other words, data analysis for the question “what influence does religious affiliation in adolescence have upon adult educational attainment?”) W r i t i n g S o c i o l o g y Findings: P a g e | 33 Being raised conservative Protestant appears to have the largest negative effect on years of schooling completed, while Jewish affiliation has the largest positive effect on the same. New Questions: Could this be partly explained by the regional differences between Conservative Protestants (who are overrepresented in the South) and Jews (who are more likely to live in the Northeast)? Could the difference in general wealth between these two regions of the U.S. partly explain this outcome? Next Steps: I should code for region at age 16 and see if this diminishes the effect of religious traditions. The same techniques are essential for analyzing qualitative data. If you’re using field observation in your research, for instance, you should include these steps in the reflecting that you do upon your growing collection of field notes. If you’re conducting interviews, you should be reading through your transcripts regularly (or just listening to recordings of your interviews) in order to begin noticing patterns, new questions, and unexpected findings in your data. Don’t wait to start analyzing your data until all of your interviews or field observations have been collected! In fact, thinking about “next steps” are particularly important for interviews and field work because this will alert you to things you should have asked about in your interviews or looked for in your field observations. Taking note of things you want to investigate early on will allow you to incorporate new dimensions of data collection into the later stages of your work. Tips for Early Data Analysis: Coding Whether you are using quantitative or qualitative data, your analysis will likely require you to engage in some strategies of breaking up your data into manageable pieces for analysis. We generally call this process “coding.” Put simply, coding is an analysis technique for labeling, separating, comparing, and organizing data. In quantitative data, the codes are applied through program language in an analysis program like STATA or SPSS. In qualitative data, you will need to construct and apply codes yourself as a way of keeping track of the things you want to focus on in your data analysis. In qualitative data – say, an interview transcript or field notes excerpt – coding is a response to implicit questions such as: •
•
•
•
What do I see going on here? What ideas, themes, and concepts appear, and how are they related to each other? What do the words (in a text, or used by my respondents) mean in this context? What underlying assumptions are present, but perhaps not explicitly stated? Coding is essential when working with qualitative data, and can be a particularly useful technique for making your research more focused and manageable in its early stages. Coding highlights things such as conceptual categories, textual features (metaphors, symbols, tone, etc.), assumptions of a speaker or W r i t i n g S o c i o l o g y P a g e | 34 the text, and “In Vivo” codes (phrases or words taken from the text itself). You can do this by hand on a hard copy of your transcript, as comments in a Word document, or in a software program specifically designed for this purpose, such as NVivio or Atlas.TI. Whatever form you choose, coding is essential for identifying the themes and questions you will eventually want to focus on in your data analysis. Coding is divided up into two phases. Initial or open coding creates a large set of codes suggested by the data itself and connects them to small bits of data, often line by line. Open coding identifies features of data that may or may not turn out to be important, begins to highlight patterns, and can suggest questions you don’t know the answer to or hadn’t initially thought of. In later stages, focused coding draws upon patterns and themes in the initial codes to formalize categories of analysis. Focused coding draws on reflective freewriting to organize key initial codes into groups or families that will be central to the analysis. The final stage of this kind of code-­‐structured analysis is called memoing. Writing analytical memos are a key way that the researcher begins identifying patterns, analytic themes, or questions that can be answered through further analysis of the data. In early stages we call these “initial memos” and the writing may be more casual, stream of consciousness, or isolated. Later on, you’ll write more structured, analytical memos that are more formalized meditations on a particular analytical question. This is the place where the relationship between codes, themes, and your research questions begin to be elaborated. For example, this process is illustrated in the following excerpt, from the early stages of an interview project studying the experiences of retail workers at big box stores. (Note that the codes here are distinguished from the interview transcript by means of capitalization): RPM: So what kinds of things did you like about working for Wal-­‐Mart? V: I liked doing the clothes, like setting the floor plans and doing the module. EMPLOYEE ENJOYMENT: MERCHANDISE RPM: You have to tell me what the means because I don’t know much about retail. V: Floor plan, you like, you know the clothes comes in the store by season. So when Spring is coming you have such beautiful clothes coming and the floor plan change and then when Winter comes you have all this heavy stuff coming in so the floor plan changes also. You have the coats on the wall. The fleece shirts in a circle rack and then no more short pants but long pants. Then you have all these sweat pans coming in and you have to line them up so people, the customers can get it. Like they can match the clothes as they go around. You have the shirts here, the pants here and the jackets on the wall and stuff like that. MERCHANDISE; JOB RESPONSIBILITIES: ARRANGING PRODUCTS RPM: So I could find the blue sweatpants that go with the blue t-­‐shirt and the blue sweatshirt? W r i t i n g S o c i o l o g y P a g e | 35 V: Yes. That’s how the book say to set it up. Then the sweats on the wall which was really…first you would have t-­‐shirts on the wall but then when the season change then you’d have sweatshirts and sweatpants on the wall, you know. Everything just change and then you’d have t-­‐shirts and when the season change you’ll have long sleeve t-­‐
shirts so that was the thing that I liked to do. I liked changing around seasons. I liked that. JOB RESPONSIBILITIES: ARRANGING PRODUCTS RPM: So would you be the person that actually got to open the box? V: Yes. RPM: So what was that like? Were you excited to see what was inside? V: Yeah, it was good. It was exciting. Then you say, oh my goodness. I’m going to get that one. I’m going to buy that too. It was really nice working at Wal-­‐Mart. Then they moved me to ladies wear because I was doing so good in men’s wear. I was doing so good in men’s wear that I received from the district manager for me and my associate 12 good job buttons. ADMIRATION OF PRODUCTS /CONSUMERISM In these early stages of the research process, I was wondering about the different ways that retail workers found enjoyment and motivation in service positions that are, in most cases, dead-­‐end jobs. In reading through this transcript, however, I began to wonder about the significance of retail workers’ interactions with products themselves. The following is an excerpt of a memo I wrote in connection with reviewing the interview transcript: MEMO, 4/12/2008 It’s interesting to note that V spends a good bit of time here talking about her excitement surrounding the merchandise that she was responsible for arranging and maintaining. I can imagine that it would get quite boring to only be selling and setting up the same merchandise (and she mentions that “the book” tells her how to do so), but V notes that she “liked changing around seasons.” These rhythms might be something to look forward to, and a source of variation in an otherwise monotonous job. It’s also interesting that she moves to describing her excitement at the idea of also buying these same goods. It would be interesting to explore the ways in which workers also think of themselves as consumers on the retail shop floor. Does that potentially influence any dimension of how workers approach their work? Need to think about this in other interviews, too. Working with your data in this way – whether qualitative, as in the above example, or quantitative as with survey data – is an excellent strategy for analysis all throughout your independent research. Rather W r i t i n g S o c i o l o g y P a g e | 36 than waiting until the end of the data collection process to begin thinking “what does it all mean?”, savvy researchers are thinking about these questions at all stages of the research process. Freewriting in the form of memos also moves you ahead when you begin to actually write about your evidence in your research paper drafts: you’ve already begun to think about important themes, and to connect them to concrete portions of evidence. Writing about Data Analysis: Citing and Interpreting Evidence Once you are on your way to recognizing the central themes and findings in your data, you will need to find a way to communicate those findings in your paper. Keep in mind that you will have access to the entire body of that data, but will not be able to report it in its entirety to your reader. Nor would your reader want you to! Your job as the researcher is to spend time analyzing the evidence you have collected, in whatever method you have chosen, to build a larger claim that can be evaluated by your readers. In order for the reader to evaluate that claim, however, you will need to cite evidence to support it so that your readers can decide for themselves if your argument is convincing. The chart below shows some of the different forms of evidence that sociologists use, depending on method: Method Evidence Format Statistical Analysis Numerical: Frequencies, Percentages regression coefficients Tables; numbers also discussed in text Interviews Quotes from respondents Frequencies (depending on size of sample) Quoted in text; may also use tables, esp. for descriptive statistics Ethnographic Observation Field notes, descriptions of events Described/quoted in text Content Analysis Frequencies (numerical), but also Tables; numbers and quotes from documents quotes also discussed in text Regardless of your method, you will need to write the results section of your paper with the goal of using the evidence you have assembled to answer the research question you have posed at the beginning of your research project. If you are writing about interviews or ethnographic/participant observation, then you will need to cite actual quotes from your interviews or field notes as evidence in your paper. Rather than showing your reader all of your quotes about a particular theme or topic, however, your job as the writer is to summarize those themes by quoting a few pieces of data that you find most representative of these larger trends. W r i t i n g S o c i o l o g y P a g e | 37 If you are conducing statistical analysis of a survey, or using content analysis that also produces numerical data about the frequency of certain variables or patterns, then you will want to first create tables that report and summarize this numerical information, and then discuss the numbers in those tables as the evidence in the body of your paper. Regardless of the kind of evidence that you cite, you will need to be sure to interpret that evidence for the reader. Interpreting the evidence means, at a minimum, telling the reader what larger significance we should take away from the citation. Let’s take a look at a couple of examples. From Baker, Wayne. (2005). America’s Crisis of Values. Princeton: Princeton University Press, p. 96: In 1981, 47% of Americans said, “Most people can be trusted.” This increased slightly to 50 percent in 1990 and fell to only 36 percent in 1995 and 2000. This generally downward trend mirrors the findings of other opinion polls, such as the National Election Studies and General Social Survey. In general, levels of interpersonal trust do not differ by moral visions, as shown in figure 3.8. Note several important things here about the manner in which Baker discusses and interprets the above figure in the text: he calls the reader’s attention to the most important patterns in this chart, and uses numbers to do so (translating the dots on the graph into numerical frequencies – e.g. “50 percent” and “36 percent”). He also references the figure in the text by name. The figure also has a number of good characteristics – The vertical axis displays the question wording that is the outcome variable here, and the horizontal axis is clearly labeled. The Figure itself also has a label, and lists the source of the data used. When citing qualitative evidence in the body of your text, you have a number of choices. You can quote extended portions of field notes or interviews (including your own questions and the respondents’ answers), and you can also bring respondents’ quotes into your paragraphs if the interviewee’s statements make a particular quote in a compelling way. What’s essential is that your discussion of the results includes your own voice in linking together the evidence to make a broader claim! At this point in your research you may not know what that finding is, but you can make some headway by thinking W r i t i n g S o c i o l o g y P a g e | 38 about the different themes or patterns that emerged in the course of your research, and writing about how those patterns help to answer your research question. For instance, in a paper analyzing retail workers and their relational attitudes on the job, Rebekah Massengill wrote: Lauren invoked a relational lens to describe a conflict with her former boss at a large department store. After describing a quibble with her manager over scheduling, Lauren’s narrative quickly moved from a description of the problem (Lauren didn’t know when she was supposed to come to work) to an evaluation of her supervisor: “If she was wrong, she could never see it or admit it. She was convinced she gave me a schedule – it turned out that another lady had had the schedule and didn’t give it to me, so I knew that I was in the right and she was just very difficult to work for. No one had liked her there.” Lauren shortly added: L: [She was] probably the most difficult person I’ve ever worked for. R: You mentioned this story about not giving you the schedule – what made her so difficult? L: Her attitude, her expectations… I guess she just assumed… like I said, she was right regardless of if you had proof that she wasn’t. Just her whole attitude.. just very abrasive, just expecting you to make mistakes. Just… just a very high strung, abrasive, negative person. Even though the alleged grievance concerned information and organization – Lauren did not receive a prompt schedule of her work hours – the terms in which she criticized her colleague were relational: she was deficient in “her whole attitude” in being an “abrasive” and “negative person.” Lauren further underscored this relational criteria when she emphasized her negative appraisal of the manager by adding that “No one had liked her there.” Although Massengill report a good deal of data here – an extended quoting of the interview interspersed with several other quotations in the body of the paper, the analysis make the larger claim that the worker’s main conflict with the manager was not about the schedule, but about the manager’s poor relational skills. The point of this piece of evidence is to show that even when the employee’s complaint was about something technical – receiving the work schedule late – the respondent moved very quickly to criticize the supervisor’s lack of relational ethic. This is part of a larger paper that argues that retail workers approach their jobs as primarily relational exercises, not economistic or technical projects. Making Larger Claims Ultimately, your senior thesis will interpret a series of pieces of evidence to build to a larger claim that will answer the research question you posed at the beginning. You may not know what this larger claim is until very late in the writing process – so you should certainly keep writing drafts until you arrive at W r i t i n g S o c i o l o g y P a g e | 39 your larger point! But you can also help this process along through some techniques proposed by Booth and colleagues in The Craft of Research. Booth, Colomb, and Williams suggest that student writers use a process called “storyboarding” to help identify larger claims in independent research. The way this process works is to use several pieces of paper to write down the claims you wish to make, and the evidence you have to support them. Putting each claim on a separate piece of paper means that it’s easy to move them around organizationally, and to look at them all at once for a visual representation of what your argument is becoming. A storyboard on three pages might look like this: Claim 1: Evidence: Claim 2: Evidence: Claim 3: Evidence: The benefit of this strategy is that you can make an infinite list of claims, see how much evidence you have for each one, and then think strategically about how you can synthesize those claims into a larger argument in your independent work. You can also look to see if there are some claims that have more evidence than others – perhaps you have enough evidence to support one central claim in your paper. Or maybe some claims that you thought would be really important don’t have enough evidence, and something else is turning out to be important. The other strength of this strategy is that it forces you to think about your claims in conjunction with the evidence you have to support them. Regardless of the source of that evidence – e.g. numbers, quotes, field notes, etc. – you must have evidence to support your claims! If you have a hunch that a certain claim is valid, but have no evidence, you can mention this in your Discussion and Conclusion. And you can certainly mention this in the text – and if your research is still in progress, be thinking all the while about the kind of data you will need to find in order to support that claim. W r i t i n g S o c i o l o g y P a g e | 40 Chapter 7: Discussing Your Findings and Drawing Conclusions The final component of your senior thesis will be a Discussion and Conclusion. This will likely be the final chapter (see Chapter 8 for a more detailed discussion of the format). These sections of the paper are devoted to summarizing the most important take-­‐away points of your research: what are the most important findings of your research, and why do they matter for the larger discipline of sociology? This section of your thesis is also the appropriate place to discuss any limitations of your study, as well as the practical applications of your research for understanding the real world, or informing public policy. Use this final section of your paper to consider some of the following questions: •
•
•
•
•
•
•
How do my findings expand, challenge, or complicate the existing literature discussed earlier in the paper? What alternative explanations are there for these findings? How and why could these be put to rest? What are any potential limitations of my study/data/sample, and how might those limitations affect my interpretations? What kinds of data or additional studies could build on your research, or improve your argument? Based on my findings, what would one expect to find in similar or future research? How might my findings change the way that researchers in this field think about this subject, and how might future research build on mine? What real world applications do my findings have? What course of action do they suggest for public policy? As you conclude your independent work, keep in mind that you don’t want to end the thesis with objections, caveats, or disclaimers that completely undermine your conclusions. All social research has some limitation or blind spot, but these shortcomings do not completely eradicate the contributions of our work. Instead, we want to be mindful of the ways in which additional research can shed light on the dimensions of our research that are beyond the scope of investigation in a particular project. End on a strong note, emphasizing how your findings pave the way for future research, and contribute to the larger enterprise of social science. W r i t i n g S o c i o l o g y P a g e | 41 Chapter 8: Bibliography Last but certainly not least, all academic writing must include a Bibliography or Works Cited page. What’s the difference? A Bibliography would include all of the works you cite in your research proposal, as well as other relevant works that you consulted in your research but may not have cited directly in your writing. A Works Cited page is restricted to scholarly work that you cite in your paper. Sociologists use varied citation formats in their writing depending on the particular style of citation requested by the academic journal in which their work is published. For the purposes of our department, however, you will use the format required by the American Journal of Sociology (AJS). Here are some examples of how you would cite different kinds of scholarly sources: Note that all of the following examples contain the same key information: all authors, the title of the work, where it was published, and for academic journals, the journal volume number (note that in AJS format you need not list the issue number). An article with one author: Gibbon, James. 2008. “God is great, God is good: Teaching god concepts in Turkish Islamic sermons.” Poetics 36: 389-­‐400. An article with multiple authors: Small, Mario Luis, Erin M. Jacobs, and Rebekah Peeples Massengill. 2008. “Why Organizational Ties Matter for Neighborhood Effects: A Study of Resource Access through Childcare Centers.” Social Forces 87: 387-­‐414. [Note: In AJS format, the convention is to cite the first author as Last Name, First Name, followed by the subsequent authors as First Name Last Name, as in the above example. Follow this format for other coauthored books, edited books and chapters, and conference presentations, etc.] A book: Cohen, Lizabeth. 2003. A Consumer's Republic: The politics of mass consumption in postwar America. New York: Vintage Books. A book section or chapter from an edited volume: Moreton, Bethany E. 2006. “It came from Bentonville: The Agrarian origins of Wal-­‐Mart culture.” Pp. 57-­‐82 in Wal-­‐Mart : The face of twenty-­‐first century capitalism, edited by Nelson Lichtenstein. New York: New Press. W r i t i n g S o c i o l o g y P a g e | 42 An unpublished or “working paper”: Bertrand, Marianne, and Sendhil Mullainathan. 2002. “Are Emily and Brendan More Employable than Lakisha and Jamal? A Field Experiment on Labor Market Discrimination.” Working paper, University of Chicago. A newspaper article: Slevin, Peter. 2000. “Life after Prison: Lack of Services Has High Price.” Washington Post, April 24. A Government report: Hakken, Jon. 1979. Discrimination against Chicanos in the Dallas Rental Housing Market: An Experimental Extension of the Housing Market Practices Survey. Washington, D.C.: U.S. Department of Housing and Urban Development. A web or electronic document: Creamer, Robert. 2009. “If Congress Fails to Pass the Obama Health Care Plan.” The Huffington Post, July 15. Retrieved on July 15, 2009 from http://www.huffingtonpost.com/robert-­‐
creamer/history-­‐shows-­‐that-­‐democr_b_233056.html. [Note: Electronic or internet documents can be confusing, particularly because the conventions for citing them are still evolving. However, it is important that your citation include the same components used in a standard text citation: the source’s author, date, title, and where it was published (or posted – as in the hosting organization on the web). In addition, your citations for electronic documents should also include the date you accessed it, and the web link. If the page does not have any discernible date, you should use the abbreviation “n.d.” in the parentheses to indicate this. In any case however, you must include the date accessed in order to account for the fact that web postings change, are removed, altered, or deleted.] A web or electronic document with a corporate or organizational author: Bureau of Labor Statistics. 2002. “Local Area Unemployment Statistics.” Accessed March 1, 2003. http://www.blr.gov/lan/home.htm Frequently Asked Questions Of course, even with this guide (and other resources on citing scholarly literature) you will have questions about situations in which the appropriate action is not clear-­‐cut. Below are some frequently asked questions: W r i t i n g S o c i o l o g y P a g e | 43 How do I cite a web page in the text? Should I do so as a parenthetical citation in the text, or as a footnote? AJS citation format gives authors the option of citing web pages as a footnote in the body of the text – in fact, the journal states that it “prefers” authors to do this because “this allows authors to explain their use of the source, including date accessed.” Whatever your decision, the important thing to remember is to cite these electronic sources just as carefully as you do the books and articles you find on the shelves of the Firestone library. Do I need to cite the electronic resource I used to find a journal article – for example, a database like ProQuest or JSTOR? You’ve probably heard conflicting advice on this question – you’ve been told that you should cite the electronic database you used to find a journal, but probably notice that scholars’ published work rarely does so in a Works Cited page. Here’s the bottom line: if you accessed the article via ProQuest and viewed it as a pdf file (in other words, the same image of the page that would have appeared in a hard copy of the journal) then you do not need to reference ProQuest or give an internet link in your citation. However, if you accessed the journal article as an html file (in other words, no pagination, as a continuous stream of text) then you should cite the database and give an access date plus the web link in your citation. The reason for this is that small errors can occur in preparing document text for publication in html, thus it is important to be accurate and cite the article in the format in which you read it. This also means that if you cite directly from an article in html format (in the form of a direct quotation), you should give the paragraph number in the parenthetical citation instead of a page number. For example: (Smith 2009, para 23). I found a really useful article by reading another scholar’s work. Do I need to cite both sources? You should cite both sources in your Bibliography or Works Cited page. In your actual text, you can handle this by citing the source that most clearly informs your own research, while also noting the place in the other source in which you found it. For example: (Slevin 2000; see also Pager 2003: 498). I read an article that quotes directly from another source, and I’d like to use the same quote in my paper. Do I really need to look up the quoted source, or can I just cite the article in which I found it? So let’s say you found Mario Small’s quotation of William Whyte (1943) on p. 5-­‐6 of Villa Victoria really informative, and you’d like to quote that in your research proposal (e.g. “It is customary for the sociologist to study the slum district in terms of ‘social disorganization’ . . .”). Since you didn’t find or read Whyte independently, you will need to cite both sources in your Works Cited page – and the intellectually honest thing to do is to actually go to the library and look up the Whyte quote yourself (after all, you will want to be sure that you cite the original source accurately as well). W r i t i n g S o c i o l o g y P a g e | 44 In the text, you should note both sources in the parenthetical citation. For example: (Whyte 1943: viii, as quoted in Small 2004: 5-­‐6). Then list both sources in your Works Cited page. For More Information Of course, you may encounter many other kinds of documents in the course of your research and have questions about how to cite them in AJS format. You may also consult the “Instructions for Authors” page at the AJS website: http://www.journals.uchicago.edu/page/ajs/instruct.html, and also peruse published articles in the journal to see how other researchers have handled ambiguous situations. W r i t i n g S o c i o l o g y P a g e | 45 Chapter 9: Other Helpful Information Developing a Productive Relationship with Your Advisor Although your thesis represents your own independent research, you will be working on your research under the supervision of a faculty member you yourself select. Faculty supervision will vary based on a number of factors: the nature of your project, your advisor’s own personality and working style, and the relationship that you develop with your advisor in the course of your research. Some faculty members may be very directive in setting up a meeting schedule with you from the start, while others will be comfortable with a more hands-­‐off, intermittent schedule of meeting and advising. There is no one “right” way to have a productive relationship with your advisor; however, there are number of things you can do to get this relationship started on the right foot. Most importantly, you should talk early on with your advisor about how he or she prefers to work with undergraduate students on their independent work. To find an advisor, you should send an email to him or her introducing yourself and requesting an initial meeting to discuss your research ideas. It’s okay if you don’t yet know your exact topic and questions, and you shouldn’t let this uncertainty keep you from making early contact with your advisor. It’s entirely appropriate that the course of your thesis research might be shaped in key ways by your early meetings and conversations with your advisor. You can use an early meeting with your advisor both to feel out the kind of working relationship that will be most productive for both of you, and get a good start on your research. For instance, you might send your advisor an email that goes something like this: Subject Line: Thesis Meeting Dear Professor X, My name is Johnetta Q. Yalie and I would like you to be my advisor for my senior thesis in sociology. Although I’m in the early stages of thinking about my research, I am most interested in questions that have to do with racial inequality in educational outcomes among college students. I know that this is an area of interest for you as well, so I am very excited to talk with you about my ideas and hear your suggestions about some of the different directions this research might pursue. Would it be possible for me to meet with you sometime this week or next to introduce myself and talk about some of my ideas? The best times for me are Monday and Wednesday mornings, or Tuesday and Thursday after 2:45 pm. I look forward to meeting you soon, and working with you during this coming year. Sincerely, Johnetta W r i t i n g S o c i o l o g y P a g e | 46 For this first meeting, you don’t need to have your thesis plans set in stone – your advisor’s job is to help direct your work in a way that leads you toward a strong original contribution to your field of study. However, you should go into the meeting prepared to talk about your interests, and at least the general topic that you are thinking about exploring in your research. If you’ve thought about the kind of methods you want to use – quantitative analysis of existing data, ethnography, an original survey, etc. – you can discuss this as well. Come prepared to take notes, ask questions, and make a good first impression by being prepared and engaged. In addition, you should ask the following during that initial meeting: How often do you prefer to meet with your thesis advisees? In your experience, what makes for a productive relationship with an advisee? Do you want to see drafts of my work in advance of the official department deadlines? How far ahead do you prefer to schedule meetings with students? Before you leave the meeting, you should plan for when you will next be in touch, and discuss the work that you will do prior to that meeting. If your advisor mentions books or articles you might find useful to read, or other faculty you might consult on your topic, you should do these things promptly. •
•
•
•
Throughout the research process, it is important to stay in touch with your advisor. If you find you are stuck in your independent work, or worried about your progress in your research, you can seek out help from various sources on campus (see the following section for other suggestions), but you shouldn’t avoid your advisor! While it’s true you don’t want to have a series of aimless conversations with the professor assigned to supervise your research, you should seek out his or her input throughout the research process, and go to the meetings prepared to ask pointed questions and take careful notes on the feedback your professor offers you. With a good relationship with your advisor, the process of independent research can be one of the most rewarding endeavors you undertake at Yale. Planning Your Work Your thesis will almost certainly be the biggest academic project you have ever worked on independently. These projects require careful planning and strategic work throughout; it goes without saying that these are not projects that can be conceived in the days or even weeks leading up to the final deadline for your written work. You will need to plan your work carefully throughout the fall and spring semesters so that your final paper meets the requirements of independent work for the Yale sociology department. The department has a set of firm deadlines that are designed to keep your work on track. There is no hard and fast rule for planning your work, but many students find it helpful to meet with their advisor when they are brainstorming about a portion of the thesis, but have not yet started writing. Leaving ample room between this meeting and the due date allows you plenty of time to address any suggestions or concerns articulated by your advisor. Therefore, it is advisable to use these dates as a way of planning your meetings with your advisor as well as your own writing schedule. W r i t i n g S o c i o l o g y P a g e | 47 For instance, you might decide that it would be useful to consult with your advisor before submitting the prospectus for your senior thesis to confirm that you are both on the same page about how your work is taking shape. You can use this goal to begin working back from key deadlines to plan your work. Departmental Deadlines for Senior Theses (See also the syllabus for SOCY 493, from which these guidelines are excerpted): •
September 30: Abstract submitted in LaTeX format by email. 300 Words Maximum •
October 28: Data, Methods, Setting Description submitted in LaTeX format by email. 5 double-­‐
spaced pages. •
December 8: Literature Review submitted in LaTeX format by email. 15 to 20 double-­‐spaced pages. •
January 27: Revision of Literature Review submitted in LaTeX format by email •
February 10: Data and/ or Theme Visualizations in Beamer format by email •
February 24: Research Posters (coding and images) due in Beamer •
March 30: Senior Thesis is due to your advisors for final approval •
April 13: Electronic copies of your thesis due to [email protected] and the DUS •
April 29: Oral presentation at the Sociology Undergraduate Research Symposium (invite your advisors and your friends) PLEASE NOTE: * UNLESS AN EXTENSION HAS BEEN GRANTED BY THE DEPARTMENTAL REPRESENTATIVE, STUDENTS WHO FAIL TO TURN IN THEIR WRITTEN WORK BY THE PRESCRIBED DUE DATE WILL BE PENALIZED ONE-­‐THIRD OF A LETTER GRADE ON THEIR FINAL INDEPENDENT-­‐WORK GRADE. **STUDENTS WHO FAIL TO TURN IN THEIR FINAL WRITTEN WORK BY APRIL 12 WILL BE PENALIZED ONE-­‐THIRD OF A LETTER GRADE FOR EVERY 48 HOURS (OR FRACTION THEREOF) OF UNEXCUSED LATENESS, WEEKEND DAYS INCLUDED. The Finished Product One question that students often ask is about the difference between a term paper and a senior thesis. The main difference concerns the length and scope of each project, along with the depth of data collection and analysis associated with each form of independent research. For instance, a student who W r i t i n g S o c i o l o g y P a g e | 48 is interested in studying megachurches (very large religious congregations) might carry out a term paper project that would review relevant literature on contemporary religious organizations, and interview 12 or 15 members of a nearby megachurch about their motivations for participating in this kind of religious organization. Comparatively, a senior thesis that explored this same topic might situate the analysis in both the sociology of religion and the sociology of organizations literatures, comparing two megachurches through 12-­‐15 interviews at each congregation, as well as field observations collected over a period of a few months. In the case of the term paper, the student would be able to draw conclusions primarily about the range of motivations that lead individuals to join megachurches, and contribute primarily to the literature in the study of religion. For a senior thesis, the project might be contributing to the scholarly literature in both religion and organizations through a more comparative study, and would involve collecting more data over a longer period of time. While there is no formal length requirement for the senior thesis, past experience suggests that a typical senior thesis is typically about twice that long – 20,000 words or 50-­‐100 pages. Of course, a paper might be shorter than this but still a strong piece of work, and the reverse is true as well: a lengthy, but rambling and unfocused senior thesis might exceed well over 100 pages and still represent a weak piece of scholarship. For this reason, you should use length as a proxy for the strength and focus of your argument: a paper that is significantly shorter than these guidelines may well have an argument that is not sufficiently complex or developed, and a paper that is substantially longer may contain an argument that is poorly-­‐organized, tangential, or otherwise lacking in concentration. In terms of the final form of the thesis, the length and structure of the paper will vary, but it should include all of elements discussed in this handbook, in some format: •
•
•
•
•
•
•
•
Your research question A statement of why your research question is important A literature review of relevant social scientific literature Description of the data you have selected for analysis A discussion of the methods used for the data analysis Analysis of results Discussion of the significance of the analysis, including general conclusions A Bibliography of works cited Of course, these are the pieces of a typical sociological research article, of the sort that you have likely read in sociology classes. How these pieces get communicated in written work will vary, but a good piece of independent work will include careful attention to all of these elements. The Senior Thesis The senior thesis is typically divided into distinct chapters. Again, no one formula need be used in all cases, but the following is an example of how students might typically structure the writing of the senior thesis. I. Front Matter W r i t i n g S o c i o l o g y P a g e | 49 The beginning of the final copy of your thesis should include the following elements: •
A Title Page: Title Student’s Name A thesis presented in partial fulfillment of the requirements for the degree of Bachelor of Arts Department of Sociology Yale University Year of graduation •
•
A Table of Contents (including page numbers) Acknowledgements (who do you wish to thank for their help in this research? Now is your chance to name names!) II. Introductory Chapter(s) The first chapter(s) of the thesis should communicate the following things noted below. Note that some students will do all of the following in one, longer introductory chapter while others will choose to communicate these elements in 2 or 3 shorter chapters. Either approach is acceptable; discuss your proposed plan of writing with your faculty advisor: •
•
•
A general overview of the research project, including the research question, its importance, and a brief description of the project’s larger findings and implications A review of relevant literature Discussion of data to be analyzed, how it was collected, and the method(s) or analysis used. III. Empirical Chapter(s) W r i t i n g S o c i o l o g y P a g e | 50 These chapters are the true “guts” of your thesis – they are where you analyze the data you’ve selected for your project, and we call them “Empirical Chapters” for this reason. Here you are not discussing the research conducted by others, but analyzing your own data to make a larger claim or argument. Most senior theses will have two empirical chapters, although some students will have only one while others will have three or more. Again, the approach you take should match your goals for your research, and should be discussed with your faculty advisor. However, most students will find that it is difficult to discuss all of your findings in one empirical chapter, so students will divide their empirical analysis into separate chapters either: •
•
•
Thematically (for instance, focused around the analysis of different themes in interview data) Methodologically (perhaps you used statistical data for one part of your thesis, and content analysis for another body of data), or Theoretically (answering one set of research questions in one case, and another set of questions in a separate chapter) IV. Discussion and Conclusion The final portion of your thesis will attempt to summarize what your research means for the discipline of sociology. What are the most important findings to take out of your research? How do they respond to the literature that you discussed in the initial literature review? What larger conclusions are suggested by your findings? What is left unanswered, for future researchers to tease out and explore? How much of this you save for the last chapter is variable. Some thesis writers will answer some of the above questions at the end of each empirical data analysis chapter; others will wait to address these “what does it all mean?” kind of questions in the last chapter of the thesis. In either case, your last chapter is your final opportunity to tell your reader why your thesis was worth reading, and remind us why your research was important! V. End Matter Here is where you’ll include the following elements: •
•
Bibliography Appendices (Including an Appendix is strictly optional. However, if you choose to include a Appendix this is where you would insert materials relevant to your research design or collection that didn’t make it into the text, for instance: survey questions, interview guide questions, or tables with quantitative results that a reader might be interested in, but weren’t germane enough to your analysis to be discussed in your text. You might also want to include in the Appendix additional background information on your research methods or theoretical issues raised by your research that a particularly interested reader might want to know about.) W r i t i n g S o c i o l o g y P a g e | 51 How to Produce this Format in LaTeX Copy and paste this template into TeXworks. Enter the text of your thesis into the appropriate fields. \documentclass[12pt]{report}
\usepackage[american]{babel}
% Search on CTAN for documentation for all of these packages (i.e., "latex package nnnn."
% Many of them provide excellent commands for formatting different aspects of the text.
% csquotes in particular provides a wealth of options, especially for formatting quotations
% from other languages.
% FORMATTING THE TITLE
\usepackage{titlesec} % Sectioning commands for titles.
\titleformat{\section}{\normalsize\scshape\centering}{\thesection}{1em}{}
\titleformat{\subsection}{\normalsize\itshape}{\thesection}{1em}{}
\usepackage{appendix} % Title formatting for appendices.
% SPECIAL PACKAGES
%\usepackage{amsmath} % Uncomment if using complex math.
% FORMATTING OPTIONS
\usepackage[autopunct=true]{csquotes} % Provides commands for defining your own quotation environment.
\usepackage{setspace} % Provides commands for setting single, 1.5, and double spacing.
\usepackage{rotating} % This package will rotate text or figures any number of degrees if called in the document.
\usepackage{subfigure} % Captioning, labeling, and positioning options for small ("sub") figures.
\usepackage[margin=1in]{geometry} % Customizes page layout. The argument here sets 1-inch margins.
\usepackage{placeins} % Defines float barriers.
\usepackage{parskip} % A hack for spacing between text and subtitles.
\setlength\parindent{1em} % Sets a 15-point paragraph indent.
% URL options.
\usepackage{hyperref}
\usepackage{url}
% TABULAR ENVIRONMENTAL SETTINGS
\usepackage{longtable} % Allows tables to flow over page boundaries. Does not work in multi-column layouts.
\usepackage{dcolumn} % Provides decimal-based tabular formatting.
\usepackage{threeparttable} % A solution for providing footnotes in tables.
\usepackage{tabularx} % Provides a new column specifier (x) that stretches the table to be as wide as specified.
\usepackage{booktabs} % Uses spacing and column "thickness" to make tables render more elegantly.
% CITATION OPTIONS
\usepackage[round]{natbib} % Round gives you parentheses.
\setcitestyle{notesep={:},aysep={}} % Correct formatting for in-text citations in sociology - (Author year:page)
\bibliographystyle{ajs.bst} % Calls the bibliography style.
\begin{document}
% Title
\title{Title IX\thanks{
Insert acknowledgements. \newline
}
W r i t i n g S o c i o l o g y P a g e | 52 }
\author{
Firstname Lastname \\
\small Department of Sociology \\
\small Yale University \\
\small A thesis presented in partial fulfillment of the \\ \small requirements for the degree of Bachelor of Arts
}
\newpage
\maketitle
\doublespacing
\begin{abstract}
\noindent \normalsize
Lorem ipsum dolor sit amet, consectetur adipiscing elit. Sed quis fringilla dolor. Vestibulum nulla sapien, efficitur in
gravida a, auctor vel elit. Ut vitae blandit tellus. Nunc iaculis auctor dignissim. Nullam ac ante dapibus, lobortis elit
vitae, elementum odio. Sed interdum et est a commodo. Duis vehicula venenatis tortor nec consequat. Nam vel
tortor malesuada, efficitur dolor eget, maximus neque. Suspendisse nec viverra sem. Mauris augue eros, consequat
at felis ut, commodo bibendum quam. Aenean est nisi, consectetur sit amet lacinia in, congue id nisl. Praesent ut
mollis turpis. Proin iaculis nulla vitae tristique eleifend.
Vestibulum ut massa eu ipsum consectetur ultrices eu eget lectus. Vivamus at congue orci. Morbi iaculis metus arcu,
quis suscipit nisi porta semper. Nam pulvinar odio et dui sodales, ac auctor turpis varius. Mauris nibh enim, sagittis
sit amet urna vel, iaculis tincidunt turpis. Proin consectetur metus non neque vehicula hendrerit. Aliquam tellus
nulla, fringilla sed vulputate mollis, ultrices at dolor. \\
Keywords: Aliquam fermentum hendrerit leo non imperdiet.\\
\end{abstract}
\newpage
\section*{Table of Contents}
~~~~~~~~~Introduction............................................................................................2-5
\newpage
Fusce nec risus eu leo accumsan viverra. Praesent tincidunt laoreet ante sit amet convallis. Phasellus semper
fermentum augue, non rutrum tortor efficitur ac. Sed nec nibh vitae purus vulputate venenatis id at erat. Fusce
blandit iaculis maximus. Vivamus sed vulputate neque, bibendum dapibus mi. Quisque aliquam elementum lacus,
nec condimentum lectus consectetur suscipit. Vestibulum laoreet gravida tortor, nec consequat orci placerat non.
Suspendisse dictum sit amet diam sed semper. Cras fermentum bibendum tempus.
Nulla facilisi. In hac habitasse platea dictumst. Vivamus dapibus, nunc ac egestas venenatis, augue sapien commodo
libero, sed feugiat ligula sapien ut orci. Vestibulum ultricies tortor nec viverra laoreet. Pellentesque egestas turpis
nisi, eget varius metus facilisis non. Fusce nec feugiat augue, a viverra elit. Morbi eu ornare mauris. Nam dapibus
lacinia lorem, eu maximus orci scelerisque nec. In hac habitasse platea dictumst. Proin tincidunt lobortis venenatis.
Duis consectetur tincidunt rutrum. Sed gravida lectus eros, lacinia interdum sapien mattis placerat. Cras ut metus sit
amet massa rutrum porttitor ac vel purus. Ut non ullamcorper tellus. Aliquam eget felis diam.
\newpage
\section*{Literature Review}
Cum sociis natoque penatibus et magnis dis parturient montes, nascetur ridiculus mus. Donec feugiat auctor quam,
ut varius nisi bibendum quis. Ut vestibulum turpis et erat tincidunt gravida. Praesent mollis a lectus viverra varius.
Morbi bibendum elementum libero vitae consequat. Duis id facilisis nibh, in elementum tellus. Suspendisse euismod
metus eros, congue volutpat tortor blandit ut. In a arcu pulvinar nulla tempus maximus suscipit ac arcu. Aliquam ut
W r i t i n g S o c i o l o g y P a g e | 53 nisl quis sem laoreet ornare pharetra eget nulla. Nulla vitae lorem sit amet quam semper mattis et ut quam. Interdum
et malesuada fames ac ante ipsum primis in faucibus. Fusce nec bibendum nisl, eu consectetur nunc. Morbi ut quam
at ipsum vestibulum blandit vel at tellus. Fusce condimentum rhoncus lacus sed vehicula. Maecenas egestas tristique
dolor, et ultricies neque consectetur nec. Fusce venenatis volutpat tincidunt.
Nunc cursus, justo quis euismod blandit, diam diam consequat dui, at accumsan mi metus at tortor. Suspendisse
potenti. Aliquam leo metus, placerat sed eros ut, eleifend luctus libero. Donec consequat dolor in aliquam convallis.
Curabitur enim elit, sollicitudin non diam nec, viverra accumsan velit. In mollis mattis sagittis. Ut non metus justo.
Etiam porta vitae dolor eget molestie.
Pellentesque habitant morbi tristique senectus et netus et malesuada fames ac turpis egestas. Vivamus sit amet ante
vehicula, vehicula arcu nec, lobortis tellus. Nunc molestie dui gravida, accumsan lacus ut, elementum urna.
Pellentesque cursus iaculis odio, nec faucibus massa imperdiet maximus. Vivamus eget tristique massa. In at
pulvinar justo, eu dictum urna. Pellentesque interdum congue justo, quis tristique ante eleifend a.
Morbi eu lorem pharetra, faucibus orci eget, sodales nulla. Etiam leo quam, viverra ut tempor laoreet, convallis
ultrices lectus. Fusce orci ante, bibendum vel vehicula ac, fringilla eget risus. Ut imperdiet, mauris ultrices
elementum facilisis, ligula erat dictum felis, quis cursus dui diam sed nulla. Duis accumsan purus nec sapien
elementum accumsan. Integer a venenatis est. Vivamus ullamcorper malesuada arcu, et mattis velit consequat
lacinia. Sed pellentesque porta enim, sit amet eleifend eros porttitor at. Donec scelerisque eros vitae gravida cursus.
\subsection*{Gender}
Donec eget iaculis magna. Etiam scelerisque dapibus ante, nec venenatis ipsum tristique vel. Vivamus nec est vitae
tellus vulputate egestas. Cras imperdiet vel sem et congue. Nam quis volutpat sapien. Sed massa velit, venenatis et
augue nec, dapibus elementum lorem. In et leo tellus. Nulla quis dui id orci viverra varius. Sed non lacinia ex. Ut
mattis malesuada quam, at pulvinar tortor pharetra at. In mollis aliquet massa, nec ornare ipsum gravida sit amet. Ut
luctus arcu eget vestibulum scelerisque.
\subsection*{Networks}
Pellentesque tincidunt sapien ac venenatis posuere. Proin risus enim, efficitur quis congue eget, luctus eu lectus. Sed
facilisis, nunc id facilisis fringilla, tellus arcu tristique justo, et maximus ipsum augue at mauris. Phasellus eu nibh
quis tellus scelerisque porta at dapibus ligula. Nulla ac aliquam erat, eu efficitur massa. Integer vel mattis tellus.
Cras libero justo, lacinia sed vulputate ut, tristique non sem. Duis pulvinar, ante nec ornare mollis, tellus ipsum
venenatis arcu, at viverra sem arcu ornare est.
\subsection*{Race and ethnicity}
Sed vel tempus tortor. Mauris sollicitudin libero nec pretium suscipit. Suspendisse cursus consequat mi. Nam
hendrerit magna eu sagittis rhoncus. Ut accumsan erat ac felis sagittis, id lobortis tellus auctor. Nulla lobortis nisl
neque, sit amet porttitor nibh luctus in. Pellentesque eu lectus nunc. Morbi at enim vitae nunc fermentum aliquam
nec ut leo.
\newpage
\section*{Data and Methods}
Curabitur \citet{Beard1986} congue nulla id ante porttitor, quis scelerisque sapien ullamcorper. Nulla maximus
sapien nisi, eget ultrices odio egestas in. Quisque id rutrum sem. Nulla quis imperdiet tortor. Class aptent taciti
sociosqu ad litora torquent per conubia nostra, per inceptos himenaeos. Curabitur volutpat auctor urna elementum
mollis. Phasellus porttitor ligula odio. Vivamus iaculis quis sem sit amet rhoncus. Sed odio quam, hendrerit ut
laoreet vel, accumsan eget est. Vestibulum lorem sapien, pharetra sit amet volutpat vel, sagittis ut velit. In hac
habitasse platea dictumst. Nullam vel interdum justo. Aenean est felis, volutpat non ipsum a, facilisis lacinia nulla.
Aliquam eros dolor, vulputate lacinia lobortis at, placerat sollicitudin dolor. Nulla auctor pharetra augue a molestie.
\subsection*{Data}
Aenean leo sapien, lacinia vel nulla at, fringilla ullamcorper mi. Pellentesque rhoncus at leo eget pharetra. Ut augue
lacus, consequat et finibus ut, tristique at nisl. Etiam vulputate tellus sed malesuada auctor. Pellentesque ac nisi
mollis, ornare ex eget, volutpat elit. Curabitur eget sem eu enim eleifend dapibus. Duis tempor, leo nec molestie
congue, diam augue bibendum magna, sed eleifend sem nulla bibendum ligula. Ut auctor condimentum libero.
W r i t i n g S o c i o l o g y P a g e | 54 Maecenas mollis vestibulum metus et pellentesque. Donec pretium libero sed lacus tempor, ut varius dui aliquet.
Quisque placerat lacus eget posuere posuere. Aliquam enim velit, lacinia egestas rutrum id, varius id odio. Nunc
vulputate felis tellus, sodales lobortis magna accumsan at. Quisque quis nisi nisl. Suspendisse sit amet tincidunt elit.
\paragraph*{Perhaps you want a subsubsection here.}
\subsection*{Methods}
Proin eu lectus convallis nulla facilisis feugiat. Duis efficitur eros quis blandit egestas. Donec a egestas arcu. In
ullamcorper enim eu pulvinar dictum. Aliquam pretium ante lectus, id mattis dui faucibus sed. Aenean quis tortor
vitae mauris viverra posuere at vitae tellus. Proin tincidunt tempus augue et finibus. Etiam interdum vel arcu nec
cursus.
Aliquam lorem neque, porttitor nec sapien et, lobortis interdum lacus. Vivamus mi ipsum, vestibulum id posuere sit
amet, laoreet eget nibh. Praesent vel egestas neque, eu lacinia ante. Quisque non leo ex. Morbi sit amet erat tempus,
egestas mauris at, imperdiet elit. Quisque sed ex eu est consectetur scelerisque nec eget nisi. In hac habitasse platea
dictumst \citep[41]{Everitt2011}. Phasellus tempus dolor vel risus egestas, ut congue tellus condimentum. Mauris
sit amet dictum nisl. Nam placerat, tellus a ornare suscipit, magna eros varius felis, vitae vulputate elit libero sit
amet ex. Donec in rhoncus augue, at varius velit. Duis rhoncus est sed ullamcorper condimentum. Sed vel sodales
enim, in porttitor urna.
Aliquam dapibus, velit eu tempus tristique, odio mi pharetra mi, ut interdum dolor est a orci. Aenean facilisis sagittis
laoreet. Proin ligula orci, semper ut ornare vitae, lobortis eu lectus. Aenean quis leo bibendum, vestibulum ligula
sollicitudin, molestie massa. Sed placerat est dui, vitae suscipit libero finibus ut. Nunc convallis ex arcu, a
sollicitudin mi blandit in. Fusce sed massa et velit luctus cursus. Vestibulum eu sem augue. Phasellus non placerat
magna. Aenean fringilla vulputate orci, at pretium mauris faucibus non. Nunc at rhoncus sem, sed efficitur tellus.
Nam sodales tempus libero quis tristique.
\newpage
\section*{Results}
Ut mollis odio interdum ipsum convallis faucibus. Suspendisse volutpat tempus augue, eu dictum augue dignissim
ut. Maecenas commodo ipsum at velit ullamcorper, pulvinar tincidunt ipsum scelerisque. Nulla luctus, enim
consequat scelerisque facilisis, ligula urna elementum turpis, eget mollis ligula elit vitae dolor. Donec eget lacus
ante. In vestibulum lorem vitae aliquam elementum. Sed convallis commodo blandit. Nam pulvinar metus et
pulvinar dignissim. Ut eget massa convallis, maximus ligula vitae, efficitur velit. Fusce justo erat, consequat non
auctor sed, pulvinar in diam. Sed orci velit, molestie ac gravida nec, rhoncus ut nibh. Vivamus aliquet leo in nibh
rhoncus sagittis \citep{Cicero1877}.
Curabitur non gravida urna. Fusce tellus massa, congue non ultricies vel, vulputate non diam. Class aptent taciti
sociosqu ad litora torquent per conubia nostra, per inceptos himenaeos. Etiam magna justo, efficitur nec euismod
vitae, auctor sed mi. Aliquam justo eros, aliquet sit amet consequat non, egestas vitae ex. Aliquam porta, risus ac
aliquet malesuada, est libero scelerisque nulla, eu convallis elit nisl id lectus. Cras euismod turpis ligula, non viverra
elit auctor nec. Vestibulum ante ipsum primis in faucibus orci luctus et ultrices posuere cubilia Curae; Integer at
finibus lacus. Phasellus facilisis dignissim tellus.
\newpage
\section*{Discussion and Conclusion}
Nam mattis libero nisi. Pellentesque vehicula tristique augue. Sed sed tristique urna, aliquam ultricies felis. Praesent
malesuada tellus ac maximus ullamcorper. Donec laoreet enim quam. In pharetra ligula a metus rhoncus, nec
interdum metus sagittis. Sed consequat odio eget commodo mattis. Phasellus dapibus mauris sit amet mi lobortis, et
blandit quam interdum. Donec id augue auctor, sagittis ligula eget, tristique felis. Sed tempor mi blandit neque
venenatis porta. Donec vel nulla leo. Fusce metus nisi, auctor ut elit quis, vehicula ultricies nunc. Proin commodo
justo eget leo iaculis sollicitudin a quis magna.
\bibliography{sample.bib}
\end{document}
W r i t i n g S o c i o l o g y P a g e | 55 More LaTeX Resources A helpful Beamer guide The complete LaTeX guide Another Beamer guide Grading Criteria Although your senior thesis will be advised by one faculty member, all senior theses will be read by another member of the faculty, called a “second reader.” They will read your final paper and in the case of the senior thesis provide written comments, as well as help determine your final grade. Below are the sociology department’s grading standards for independent work: A. Normally independent work in the A range should present an argument or propose an answer to a well-­‐defined and significant question or set of questions, and it should indicate with care and accuracy the import of its subject for the understanding of sociology. Such independent work will have substantial elements of originality in its conception of its subject, in the evidence and reasoning it brings to bear on that subject, in the analytical techniques it employs, or in all of these; even a specialist in its field of study will find that it contributes to his/her understanding. Independent work in the A range must be grounded in systematic research appropriate to its scope and objectives. Such research will almost always involve attention to the important works on the subject and may require the consultation of original documents, compilation of statistical data, or interviewing and/ or fieldwork. To merit an A, independent work should be well written, that is, it should develop its subject in an orderly way and present its ideas clearly and crisply. Poor grammar and style and more than occasional misspellings have no place in independent work receiving an A. The mark of A+ should be reserved for independent work that satisfies all of these criteria in high degree. The mark of A-­‐ should be given independent work which shows originality but does not meet in a fully satisfactory way one or two other of the requirements of independent work in the A range. B. Independent work in the B range is a less outstanding treatment of a significant subject. A specialist in its field of study should find it informative, though it will yield few insights of interest to a specialist. A well done case study which yields some, if few, lessons of general import, or a good critical review of a significant body of thought that does not carry one much beyond previous work on the subject, would merit a grade in this range. Like the A independent work, that in the B range should be grounded in a substantial amount of research appropriate to its objectives, but the latter will fail to do all that is required for systematic coverage or will W r i t i n g S o c i o l o g y P a g e | 56 ignore important sources. Independent work in the B range should be clearly written and logically organized. A B+ is an appropriate grade for a sensibly conceived, well-­‐executed, well-­‐written project that shows little originality. A B-­‐ is appropriate for well-­‐conceived projects that have some significant flaw in execution or a number of less important shortcomings. C. Independent work in the C range is a competent but not distinguished treatment of a significant subject. A non-­‐specialist should find it informative. It will show evidence of substantial, though not wholly adequate research, and may be flawed in one or two additional ways as well: the logic of an important argument may be faulty, the significance of findings may be explored inadequately, or the writing may be mediocre (though it must generally be clear in its expression of ideas). An informative case study that goes little beyond a narration of events, or a review of some body of literature that gets things right but does little more, should be given a grade in the C range. C+’s should be given to the most informative of independent work in the C range, C-­‐‘s to those that meet the basic requirements of the category but have several serious flaws. D. To merit the grade of D, independent work must treat a non-­‐trivial subject in sociology and must show evidence that its writer has some substantial knowledge about, and understanding of, that subject. Beyond that little can be said in praise of independent work in the D range. F. Independent work that does not meet the minimal requirements for the grade of D should be given an F. W r i t i n g S o c i o l o g y P a g e | 57 Appendix: Other Resources Yale University contains a myriad of resources designed to help you with your independent research. The final section of this handbook details some of the departments, people, and additional reading you might find helpful as you work on your thesis – both during the fall semester in SOCY 493, and beyond. University Resources and Departments Yale College Writing Center http://writing.yalecollege.yale.edu/ The Yale College Writing Center is located at 35 Broadway. It supports writers through free tutoring at the Center and in the residential colleges and through workshops on writing techniques. Tutors are available by appointment in every Residential College. For information, or to learn about drop-­‐in service at 35 Broadway, see http://writing.yalecollege.yale.edu/tutoring-­‐resources. Undergraduates who want to make an appointment can immediately visit www.yalewco.com Tutoring and Academic Support http://yalecollege.yale.edu/content/tutoring-­‐and-­‐academic-­‐support Yale undergraduates have access to a robust system of support and advice, in the form of tutors sponsored by residential colleges, academic departments, or the Yale College Dean’s Office, as well as a variety of subject-­‐focused centers and programs. To get started, visit the website for resources. Tutors and workshops are available for specific subject areas, including the social sciences and quantitative reasoning. Center for Science and Social Science Information (CSSSI) http://csssi.yale.edu/ The Center for Science and Social Science Information (CSSSI), located in the Kline Biology Tower, provides Yale faculty, students, and staff with state-­‐of-­‐the-­‐art information services in a technology-­‐rich environment. It is designed to provide easily accessible support for science, social science and interdisciplinary researchers. Librarians are available there to assist you. The Center also houses the StatLab consultants (http://statlab.stat.yale.edu/people/showConsultants.jsp) who can help with quantitative projects. W r i t i n g S o c i o l o g y P a g e | 58 Useful People and Contacts It goes without saying that some of the most important people for your research include the professor leading SOC 493, its TF, and the department advisor who is assigned to mentor your research. However, many other people throughout the university community are here to help as well. In particular, the library employs reference librarians in particular subject areas who are particularly knowledgeable about the larger literatures in their subject area, and in many cases have extended familiarity with data collections that might be useful to you in your analysis. To connect with any of these specialists, simply email them from your Yale email account! Your independent research may take you into literatures that overlap with a neighboring discipline – such as African-­‐American Studies, Psychology, or Religion. The librarians in those subject areas will be a great resource for your research. Use them! African-­‐American Studies Gary, David (203) 432-­‐5165 [email protected] Anthropology, Economics, Gay and Lesbian Studies, Gender Studies, Psychology, Sociology, Statistics, Women’s Studies Crowley, Gwyneth (203) 432-­‐3213 [email protected] Finance, Management Nonprofit Organizations, Organizational Behavior Wachowicz, Erin (203) 432-­‐3306 [email protected] Forestry & Environmental Studies Heister, Carla (203) 432-­‐5132 [email protected] History of Medicine Grafe, Melissa (203) 785-­‐4354 [email protected] History of Science Gary, David (203) 432-­‐5165 [email protected] W r i t i n g S o c i o l o g y Middle East Studies Dougherty, Roberta L (203) 432-­‐1373 [email protected] Political Science, Public Policy Maksin, Melanie (203) 432-­‐3310 [email protected] Religious Studies Estelle-­‐Holmer, Suzanne (203) 432-­‐6374 suzanne.estelle-­‐[email protected] Social Science Data & Science Data Hudson, Michelle (203) 432-­‐4587 [email protected] For Further Reading P a g e | 59 George, Mary. 2008. The Elements of Library Research: What Every Study Needs to Know. Princeton: Princeton University Press. If you’re looking for more guidance on the murky world of library research – particularly advice on the best way to use internet resources – this is a terrific book written by one of Princeton’s reference librarians. The book also has a number of helpful checklists regarding timelines in the research process, and ways to make the most use of your professors and advisors. Booth, Wayne C., Gregory G. Colomb, and Joseph M. Williams. 2008. The Craft of Research (3rd Edition). Chicago: University of Chicago Press. This is an outstanding book for academic researchers of all disciplines, and is written with the undergraduate student in mind. The authors take you through all of the steps of finding a research question, to organizing your writing and preparing your final research paper. The book also has some useful sections on using charts and figures, and communicating evidence visually in the forms of graphs and tables. Emerson, Robert M., Rachel I. Fretz, and Linda L. Shaw. 1995. Writing Ethnographic Fieldnotes. Chicago: University of Chicago Press. This book is a classic for sociologists conducting ethnographic research, and contrary to what its title would suggest, the book actually covers most aspects of ethnographic research, from the actual techniques of taking notes in the field, to writing them up and analyzing them in a written report. The W r i t i n g S o c i o l o g y P a g e | 60 books’ final chapter, “Writing an Ethnography” presents some particularly useful techniques for transforming the ideas and theories inherent in ethnographic field notes into a written report for a larger audience. If your thesis uses ethnographic methods, you will want to consult this book. Weiss, Robert W. 1995. Learning from Strangers: The Art and Method of Qualitative Interview Studies. New York: The Free Press. Weiss’ book discusses the theory, logic, and logistics of interviewing respondents in qualitative research. Not only does this volume offer a number of useful strategies for soliciting good data from your respondents, the book also includes a chapter on writing about interview data which discusses technical issues such as which quotes to select and how to quote respondents effectively. A must for students writing a thesis based on interviews. Becker, Howard S. 1986. Writing for Social Scientists: How to start and finish your thesis, book, or article. Chicago: University of Chicago Press. Writing for Social Scientists is an outstanding book that speaks to undergraduate students and seasoned professors alike about the dilemmas of academic writing and concrete strategies for overcoming them. Notable here are chapters about getting started writing “One Right Way” and preparing a literature review (“Terrorized by the Literature”). Most importantly, Becker here makes two important points: one, that writing should not be a private process, and two, that sociologists must write clearly. Becker, Howard S. 1998. Tricks of the Trade: How to think about your research while you’re doing it. Chicago: University of Chicago Press. This methodological book has five chapters that deal with different aspects of the research process, though not chronologically. Becker’s writing is entertaining, clear, and thoughtful, and above all, offers advice to sociological researchers about helpful strategies (what he terms “tricks”) for social research that probe for deeper, unconventional, even surprising findings.